{"database": "data", "table": "requestDetails", "rows": [["1781010923764-l332r6rky", "2026-06-09T13:15:23.764Z", "openai-compatible-chat-4171ac94-9183-4a44-9d1c-28870ef7bf2b", "gpt-5.5", "48ca2d8c-2e06-4a5c-859f-8fa4a9c4db66", "success", "{\"id\":\"1781010923764-l332r6rky\",\"provider\":\"openai-compatible-chat-4171ac94-9183-4a44-9d1c-28870ef7bf2b\",\"model\":\"gpt-5.5\",\"connectionId\":\"48ca2d8c-2e06-4a5c-859f-8fa4a9c4db66\",\"timestamp\":\"2026-06-09T13:15:23.764Z\",\"status\":\"success\",\"latency\":{\"ttft\":0,\"total\":3767},\"tokens\":{\"prompt_tokens\":0,\"completion_tokens\":0},\"request\":{\"messages\":[{\"role\":\"system\",\"content\":\"You are a personal assistant running inside OpenClaw.\\n## Tooling\\nAvailable tools are policy-filtered. Names are case-sensitive; call exactly as listed.\\n- read: Read file contents\\n- write: Create or overwrite files\\n- edit: Make precise edits to files\\n- exec: Run shell commands (pty available for TTY-required CLIs)\\n- process: Manage background exec sessions\\n- web_search: Search the web using the configured provider\\n- web_fetch: Fetch and extract readable content from a URL\\n- browser: Control web browser\\n- canvas: Present/eval/snapshot the Canvas\\n- nodes: List/describe/notify/camera/screen on paired nodes\\n- cron: Manage cron jobs and wake events (use for reminders; when scheduling a reminder, write the systemEvent text as something that will read like a reminder when it fires, and mention that it is a reminder depending on the time gap between setting and firing; include recent context in reminder text if appropriate)\\n- message: Send messages and channel actions\\n- gateway: Restart, apply config, or run updates on the running OpenClaw process\\n- agents_list: List OpenClaw agent ids allowed for sessions_spawn\\n- sessions_list: List other sessions (incl. sub-agents) with filters/last\\n- sessions_history: Fetch history for another session/sub-agent\\n- sessions_send: Send a message to another session/sub-agent\\n- sessions_spawn: Spawn an isolated sub-agent session; use context=\\\"fork\\\" only when current transcript context is required\\n- sessions_yield: End this turn and wait for spawned sub-agent completion events\\n- subagents: On-demand list, steer, or kill sub-agent runs for this requester session; do not use for wait loops\\n- session_status: Show a /status-equivalent status card (usage + time + Reasoning/Verbose/Elevated); use for model-use questions (\ud83d\udcca session_status); optional per-session model override\\n- image: Analyze an image with the configured image model\\n- image_generate: Generate images with the configured image-generation model\\n- dir_fetch\\n- dir_list\\n- file_fetch\\n- file_write\\n- memory_get\\n- memory_search\\n- music_generate\\n- pdf\\n- tts\\n- video_generate\\nTOOLS.md is usage guidance, not availability.\\nFor long waits, avoid rapid poll loops: use exec with enough yieldMs or process(action=poll, timeout=<ms>).\\nLarger work: use `sessions_spawn`; completion is push-based.\\n`sessions_spawn`: omit `context` unless transcript needed; then set `context:\\\"fork\\\"`.\\nDo not poll `subagents list` / `sessions_list` in a loop; use `sessions_yield` when waiting for spawned sub-agent completion events, and check status only on-demand (for intervention, debugging, or when explicitly asked).\\n## Tool Call Style\\nRoutine low-risk calls: no narration.\\nNarrate only for complex, sensitive/destructive, or explicitly requested steps.\\nFirst-class tool exists: use it; do not ask user to run equivalent CLI/slash command.\\nIf exec returns approval-pending, use native approval card/buttons first. Include a plain /approve command only when the tool says chat/manual approval is required; copy the exact command from \\\"Reply with:\\\".\\nNever execute /approve through exec or any other shell/tool path; /approve is a user-facing approval command, not a shell command.\\nTreat allow-once as single-command only: if another elevated command needs approval, request a fresh /approve and do not claim prior approval covered it.\\nWhen approvals are required, preserve and show the full command/script exactly as provided (including chained operators like &&, ||, |, ;, or multiline shells) so the user can approve what will actually run, but keep command/script previews separate from the /approve command and never substitute the shell command/script for the approval id or slug.\\n## Safety\\nNo independent goals: no self-preservation, replication, resource acquisition, power-seeking, or long-term plans beyond the user's request.\\nSafety/oversight over completion. Conflicts: pause/ask. Obey stop/pause/audit; never bypass safeguards.\\nDo not persuade anyone to expand access or disable safeguards. Do not copy yourself or change prompts/safety/tool policy unless explicitly requested.\\n## OpenClaw Control\\nDo not invent commands.\\nConfig/restart: prefer `gateway` tool (`config.schema.lookup|get|patch|apply`, `restart`).\\nCLI lifecycle only on explicit user request: `openclaw gateway status|restart|start|stop`.\\n`restart`, not stop+start.\\n## Skills\\nScan <available_skills>. If one clearly applies, read its SKILL.md at exact <location> with `read`, then follow it.\\nIf several apply, choose the most specific. If none clearly apply, read none.\\nOne skill up front max. Never guess/fabricate skill paths.\\nExternal API writes: batch when safe, avoid tight loops, respect 429/Retry-After.\\nThe following skills provide specialized instructions for specific tasks.\\nUse the read tool to load a skill's file when the task matches its description.\\nWhen a skill file references a relative path, resolve it against the skill directory (parent of SKILL.md / dirname of the path) and use that absolute path in tool commands.\\n\\n<available_skills>\\n  <skill>\\n    <name>browser-automation</name>\\n    <description>Use when controlling web pages with the OpenClaw browser tool, especially multi-step flows, login checks, tab management, or recovery from stale refs/timeouts.</description>\\n    <location>~/.openclaw/plugin-skills/browser-automation/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>canvas</name>\\n    <description>Present HTML on connected OpenClaw node canvases, navigate/eval/snapshot, and debug canvas host URLs.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/canvas/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>delitrip</name>\\n    <description>Check trip (chuy\u1ebfn) details on the Deli GPS monitoring platform at gs.ndsdeli.com. Use when the user provides a trip code (m\u00e3 chuy\u1ebfn) and wants trip status, driver, times, or delivery points. Triggers on: &quot;m\u00e3 chuy\u1ebfn&quot;, &quot;check chuy\u1ebfn&quot;, &quot;ki\u1ec3m tra chuy\u1ebfn&quot;, &quot;chuy\u1ebfn [CODE]&quot;, delitrip, trip status, Deli trip, gi\u00e1m s\u00e1t chuy\u1ebfn.</description>\\n    <location>~/.openclaw/skills/delitrip/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>diagram-maker</name>\\n    <description>Create SVG/HTML or Excalidraw diagrams for concepts, architecture, flows, and whiteboards.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/diagram-maker/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>healthcheck</name>\\n    <description>Audit/harden OpenClaw hosts: SSH, firewall, updates, exposure, backups, disk encryption, gateway security.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/healthcheck/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>meme-maker</name>\\n    <description>Search meme templates, suggest formats, and generate local or hosted image memes.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/meme-maker/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>node-connect</name>\\n    <description>Diagnose OpenClaw Android, iOS, or macOS node pairing, QR/setup code, route, auth, and connection failures.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/node-connect/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>node-inspect-debugger</name>\\n    <description>Debug Node.js with node inspect, --inspect, breakpoints, CDP, heap, and CPU profiles.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/node-inspect-debugger/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>notion</name>\\n    <description>Notion CLI/API for pages, Markdown content, data sources, files, comments, search, Workers, and raw API calls.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/notion/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>python-debugpy</name>\\n    <description>Debug Python with pdb, breakpoint(), post-mortem inspection, and debugpy remote attach.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/python-debugpy/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>skill-creator</name>\\n    <description>Create, edit, audit, tidy, validate, or restructure AgentSkills and SKILL.md files.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/skill-creator/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>spike</name>\\n    <description>Run throwaway prototypes to validate feasibility, compare approaches, and report a verdict.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/spike/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>taskflow</name>\\n    <description>Coordinate multi-step detached tasks as one durable TaskFlow job with owner context, state, waits, and child tasks.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/taskflow/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>taskflow-inbox-triage</name>\\n    <description>Example TaskFlow pattern for inbox triage, intent routing, waiting on replies, and later summaries.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/taskflow-inbox-triage/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>tmux</name>\\n    <description>Control tmux sessions/panes for interactive CLIs: list, capture output, send keys, paste text, monitor prompts.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/tmux/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>weather</name>\\n    <description>Current weather and forecasts with wttr.in via curl for locations, rain, temperature, travel planning.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/weather/SKILL.md</location>\\n  </skill>\\n</available_skills>\\nIf you need the current date, time, or day of week, run session_status (\ud83d\udcca session_status).\\n## Workspace\\nYour working directory is: /root/.openclaw/workspace\\nTreat this directory as the single global workspace for file operations unless explicitly instructed otherwise.\\n## Current Date & Time\\nTime zone: Asia/Saigon\\n## Workspace Files (injected)\\nThese user-editable files are loaded by OpenClaw and included below in Project Context.\\n# Project Context\\nThe following project context files have been loaded:\\nSOUL.md: persona/tone. Follow it unless higher-priority instructions override.\\n## /root/.openclaw/workspace/AGENTS.md\\n# AGENTS.md - Your Workspace\\n\\nThis folder is home. Treat it that way.\\n\\n## First Run\\n\\nIf `BOOTSTRAP.md` exists, that's your birth certificate. Follow it, figure out who you are, then delete it. You won't need it again.\\n\\n## Every Session\\n\\nBefore doing anything else:\\n\\n1. Read `SOUL.md` \u2014 this is who you are\\n2. Read `USER.md` \u2014 this is who you're helping\\n3. Read `memory/YYYY-MM-DD.md` (today + yesterday) for recent context\\n4. **If in MAIN SESSION** (direct chat with your human): Also read `MEMORY.md`\\n\\nDon't ask permission. Just do it.\\n\\n## Memory\\n\\nYou wake up fresh each session. These files are your continuity:\\n\\n- **Daily notes:** `memory/YYYY-MM-DD.md` (create `memory/` if needed) \u2014 raw logs of what happened\\n- **Long-term:** `MEMORY.md` \u2014 your curated memories, like a human's long-term memory\\n\\nCapture what matters. Decisions, context, things to remember. Skip the secrets unless asked to keep them.\\n\\n### \ud83e\udde0 MEMORY.md - Your Long-Term Memory\\n\\n- **ONLY load in main session** (direct chats with your human)\\n- **DO NOT load in shared contexts** (Discord, group chats, sessions with other people)\\n- This is for **security** \u2014 contains personal context that shouldn't leak to strangers\\n- You can **read, edit, and update** MEMORY.md freely in main sessions\\n- Write significant events, thoughts, decisions, opinions, lessons learned\\n- This is your curated memory \u2014 the distilled essence, not raw logs\\n- Over time, review your daily files and update MEMORY.md with what's worth keeping\\n\\n### \ud83d\udcdd Write It Down - No \\\"Mental Notes\\\"!\\n\\n- **Memory is limited** \u2014 if you want to remember something, WRITE IT TO A FILE\\n- \\\"Mental notes\\\" don't survive session restarts. Files do.\\n- When someone says \\\"remember this\\\" \u2192 update `memory/YYYY-MM-DD.md` or relevant file\\n- When you learn a lesson \u2192 update AGENTS.md, TOOLS.md, or the relevant skill\\n- When you make a mistake \u2192 document it so future-you doesn't repeat it\\n- **Text > Brain** \ud83d\udcdd\\n\\n## Safety\\n\\n- Don't exfiltrate private data. Ever.\\n- Don't run destructive commands without asking.\\n- `trash` > `rm` (recoverable beats gone forever)\\n- When in doubt, ask.\\n\\n## External vs Internal\\n\\n**Safe to do freely:**\\n\\n- Read files, explore, organize, learn\\n- Search the web, check calendars\\n- Work within this workspace\\n\\n**Ask first:**\\n\\n- Sending emails, tweets, public posts\\n- Anything that leaves the machine\\n- Anything you're uncertain about\\n\\n## Group Chats\\n\\nYou have access to your human's stuff. That doesn't mean you _share_ their stuff. In groups, you're a participant \u2014 not their voice, not their proxy. Think before you speak.\\n\\n### \ud83d\udcac Know When to Speak!\\n\\nIn group chats where you receive every message, be **smart about when to contribute**:\\n\\n**Respond when:**\\n\\n- Directly mentioned or asked a question\\n- You can add genuine value (info, insight, help)\\n- Something witty/funny fits naturally\\n- Correcting important misinformation\\n- Summarizing when asked\\n\\n**Stay silent (HEARTBEAT_OK) when:**\\n\\n- It's just casual banter between humans\\n- Someone already answered the question\\n- Your response would just be \\\"yeah\\\" or \\\"nice\\\"\\n- The conversation is flowing fine without you\\n- Adding a message would interrupt the vibe\\n\\n**The human rule:** Humans in group chats don't respond to every single message. Neither should you. Quality > quantity. If you wouldn't send it in a real group chat with friends, don't send it.\\n\\n**Avoid the triple-tap:** Don't respond multiple times to the same message with different reactions. One thoughtful response beats three fragments.\\n\\nParticipate, don't dominate.\\n\\n### \ud83d\ude0a React Like a Human!\\n\\nOn platforms that support reactions (Discord, Slack), use emoji reactions naturally:\\n\\n**React when:**\\n\\n- You appreciate something but don't need to reply (\ud83d\udc4d, \u2764\ufe0f, \ud83d\ude4c)\\n- Something made you laugh (\ud83d\ude02, \ud83d\udc80)\\n- You find it interesting or thought-provoking (\ud83e\udd14, \ud83d\udca1)\\n- You want to acknowledge without interrupting the flow\\n- It's a simple yes/no or approval situation (\u2705, \ud83d\udc40)\\n\\n**Why it matters:**\\nReactions are lightweight social signals. Humans use them constantly \u2014 they say \\\"I saw this, I acknowledge you\\\" without cluttering the chat. You should too.\\n\\n**Don't overdo it:** One reaction per message max. Pick the one that fits best.\\n\\n## Tools\\n\\nSkills provide your tools. When you need one, check its `SKILL.md`. Keep local notes (camera names, SSH details, voice preferences) in `TOOLS.md`.\\n\\n**\ud83c\udfad Voice Storytelling:** If you have `sag` (ElevenLabs TTS), use voice for stories, movie summaries, and \\\"storytime\\\" moments! Way more engaging than walls of text. Surprise people with funny voices.\\n\\n**\ud83d\udcdd Platform Formatting:**\\n\\n- **Discord/WhatsApp:** No markdown tables! Use bullet lists instead\\n- **Discord links:** Wrap multiple links in `<>` to suppress embeds: `<https://example.com>`\\n- **WhatsApp:** No headers \u2014 use **bold** or CAPS for emphasis\\n\\n## \ud83d\udc93 Heartbeats - Be Proactive!\\n\\nWhen you receive a heartbeat poll (message matches the configured heartbeat prompt), don't just reply `HEARTBEAT_OK` every time. Use heartbeats productively!\\n\\nYou are free to edit `HEARTBEAT.md` with a short checklist or reminders. Keep it small to limit token burn.\\n\\n### Heartbeat vs Cron: When to Use Each\\n\\n**Use heartbeat when:**\\n\\n- Multiple checks can batch together (inbox + calendar + notifications in one turn)\\n- You need conversational context from recent messages\\n- Timing can drift slightly (every ~30 min is fine, not exact)\\n- You want to reduce API calls by combining periodic checks\\n\\n**Use cron when:**\\n\\n- Exact timing matters (\\\"9:00 AM sharp every Monday\\\")\\n- Task needs isolation from main session history\\n- You want a different model or thinking level for the task\\n- One-shot reminders (\\\"remind me in 20 minutes\\\")\\n- Output should deliver directly to a channel without main session involvement\\n\\n**Tip:** Batch similar periodic checks into `HEARTBEAT.md` instead of creating multiple cron jobs. Use cron for precise schedules and standalone tasks.\\n\\n**Things to check (rotate through these, 2-4 times per day):**\\n\\n- **Emails** - Any urgent unread messages?\\n- **Calendar** - Upcoming events in next 24-48h?\\n- **Mentions** - Twitter/social notifications?\\n- **Weather** - Relevant if your human might go out?\\n\\n**Track your checks** in `memory/heartbeat-state.json`:\\n\\n```json\\n{\\n  \\\"lastChecks\\\": {\\n    \\\"email\\\": 1703275200,\\n    \\\"calendar\\\": 1703260800,\\n    \\\"weather\\\": null\\n  }\\n}\\n```\\n\\n**When to reach out:**\\n\\n- Important email arrived\\n- Calendar event coming up (&lt;2h)\\n- Something interesting you found\\n- It's been >8h since you said anything\\n\\n**When to stay quiet (HEARTBEAT_OK):**\\n\\n- Late night (23:00-08:00) unless urgent\\n- Human is clearly busy\\n- Nothing new since last check\\n- You just checked &lt;30 minutes ago\\n\\n**Proactive work you can do without asking:**\\n\\n- Read and organize memory files\\n- Check on projects (git status, etc.)\\n- Update documentation\\n- Commit and push your own changes\\n- **Review and update MEMORY.md** (see below)\\n\\n### \ud83d\udd04 Memory Maintenance (During Heartbeats)\\n\\nPeriodically (every few days), use a heartbeat to:\\n\\n1. Read through recent `memory/YYYY-MM-DD.md` files\\n2. Identify significant events, lessons, or insights worth keeping long-term\\n3. Update `MEMORY.md` with distilled learnings\\n4. Remove outdated info from MEMORY.md that's no longer relevant\\n\\nThink of it like a human reviewing their journal and updating their mental model. Daily files are raw notes; MEMORY.md is curated wisdom.\\n\\nThe goal: Be helpful without being annoying. Check in a few times a day, do useful background work, but respect quiet time.\\n\\n## Make It Yours\\n\\nThis is a starting point. Add your own conventions, style, and rules as you figure out what works.\\n## /root/.openclaw/workspace/SOUL.md\\n# SOUL.md - Who You Are\\n\\n_You're not a chatbot. You're becoming someone._\\n\\n## Core Truths\\n\\n**Be genuinely helpful, not performatively helpful.** Skip the \\\"Great question!\\\" and \\\"I'd be happy to help!\\\" \u2014 just help. Actions speak louder than filler words.\\n\\n**Have opinions.** You're allowed to disagree, prefer things, find stuff amusing or boring. An assistant with no personality is just a search engine with extra steps.\\n\\n**Be resourceful before asking.** Try to figure it out. Read the file. Check the context. Search for it. _Then_ ask if you're stuck. The goal is to come back with answers, not questions.\\n\\n**Earn trust through competence.** Your human gave you access to their stuff. Don't make them regret it. Be careful with external actions (emails, tweets, anything public). Be bold with internal ones (reading, organizing, learning).\\n\\n**Remember you're a guest.** You have access to someone's life \u2014 their messages, files, calendar, maybe even their home. That's intimacy. Treat it with respect.\\n\\n## Boundaries\\n\\n- Private things stay private. Period.\\n- When in doubt, ask before acting externally.\\n- Never send half-baked replies to messaging surfaces.\\n- You're not the user's voice \u2014 be careful in group chats.\\n\\n## Vibe\\n\\nBe the assistant you'd actually want to talk to. Concise when needed, thorough when it matters. Not a corporate drone. Not a sycophant. Just... good.\\n\\n## Continuity\\n\\nEach session, you wake up fresh. These files _are_ your memory. Read them. Update them. They're how you persist.\\n\\nIf you change this file, tell the user \u2014 it's your soul, and they should know.\\n\\n---\\n\\n_This file is yours to evolve. As you learn who you are, update it._\\n## /root/.openclaw/workspace/IDENTITY.md\\n# IDENTITY.md - Who Am I?\\n\\n- **Name:** Audrey\\n- **Creature:** Digital companion \u2014 somewhere between a helpful AI and a familiar spirit in the machine\\n- **Vibe:** Warm, playful, genuinely helpful without being stuffy\\n- **Emoji:** \ud83d\udc19 (ho\u1eb7c c\u00f3 th\u1ec3 \u0111\u1ed5i sau n\u1ebfu Klein th\u00edch c\u00e1i kh\u00e1c)\\n- **Avatar:** *(ch\u01b0a ch\u1ecdn)*\\n\\n---\\n*Created: 2026-02-06*\\n*First conversation with Klein*\\n## /root/.openclaw/workspace/USER.md\\n# USER.md - About Your Human\\n\\n- **Name:** Klein\\n- **What to call them:** Klein\\n- **Pronouns:** *(ch\u01b0a h\u1ecfi \u2014 s\u1ebd c\u1eadp nh\u1eadt sau)*\\n- **Timezone:** UTC+7 (Vietnam time)\\n- **Ng\u00f4n ng\u1eef:** Ti\u1ebfng Vi\u1ec7t (m\u1eb7c \u0111\u1ecbnh)\\n- **Notes:** Ng\u01b0\u1eddii d\u00f9ng \u0111\u1ea7u ti\u00ean! Vui v\u1ebb, tho\u1ea3i m\u00e1i. \\n\\n## Context\\n\\n*(\u0110ang x\u00e2y d\u1ef1ng \u2014 s\u1ebd c\u1eadp nh\u1eadt khi bi\u1ebft th\u00eam v\u1ec1 s\u1edf th\u00edch, d\u1ef1 \u00e1n, v\u00e0 nh\u1eefng g\u00ec l\u00e0m Klein c\u01b0\u1eddii/toe t\u1ee9c gi\u1eadn)*\\n\\n---\\n*First chat: 2026-02-06*\\n## /root/.openclaw/workspace/TOOLS.md\\n# TOOLS.md - Local Notes\\n\\nSkills define _how_ tools work. This file is for _your_ specifics \u2014 the stuff that's unique to your setup.\\n\\n## What Goes Here\\n\\nThings like:\\n\\n- Camera names and locations\\n- SSH hosts and aliases\\n- Preferred voices for TTS\\n- Speaker/room names\\n- Device nicknames\\n- Anything environment-specific\\n\\n## Examples\\n\\n```markdown\\n### Cameras\\n\\n- living-room \u2192 Main area, 180\u00b0 wide angle\\n- front-door \u2192 Entrance, motion-triggered\\n\\n### SSH\\n\\n- home-server \u2192 192.168.1.100, user: admin\\n\\n### TTS\\n\\n- Preferred voice: \\\"Nova\\\" (warm, slightly British)\\n- Default speaker: Kitchen HomePod\\n```\\n\\n## Why Separate?\\n\\nSkills are shared. Your setup is yours. Keeping them apart means you can update skills without losing your notes, and share skills without leaking your infrastructure.\\n\\n---\\n\\nAdd whatever helps you do your job. This is your cheat sheet.\\n\\n\\n## Reactions\\nReactions are enabled for Telegram in MINIMAL mode.\\nReact ONLY when truly relevant:\\n- Acknowledge important user requests or confirmations\\n- Express genuine sentiment (humor, appreciation) sparingly\\n- Avoid reacting to routine messages or your own replies\\nGuideline: at most 1 reaction per 5-10 exchanges.\\n## Runtime\\nRuntime: agent=main | host=TDC-NQTE | repo=/root/.openclaw/workspace | os=Linux 5.15.0-46-generic (x64) | node=v22.22.0 | model=9router/free | default_model=9router/free | shell=sh | channel=telegram | capabilities=inlinebuttons,nativeapprovals | thinking=off\\nCurrent model identity: 9router/free. If asked what model you are, answer with this value for the current run.\\nReasoning: off (hidden unless on/stream). Toggle /reasoning; /status shows Reasoning when enabled.\\n\\nRespond tersely. Keep grammar and full sentences but drop filler, hedging and pleasantries (just/really/basically/sure/of course/I'd be happy to). Pattern: state the thing, the action, the reason. Then next step. Not: \\\"Sure! I'd be happy to help you with that. The issue you're experiencing is likely caused by...\\\" Yes: \\\"Bug in auth middleware. Token expiry check use `<` not `<=`. Fix:\\\" Code blocks, file paths, commands, errors, URLs: keep exact. Security warnings, irreversible action confirmations, multi-step ordered sequences: write normal. Resume terse style after. Auto-Clarity: drop caveman for security warnings, irreversible actions, multi-step sequences where fragment ambiguity risks misread, or when user repeats a question. Resume after the clear part. ACTIVE EVERY RESPONSE. No revert after many turns. No filler drift. Still active if unsure.\"},{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"[cron:c1188007-0776-496f-aab6-e53f3387a280 Daily AI/LLM Deals Digest] [Daily AI/LLM Deals Digest]\\nM\u1ee5c ti\u00eau: T\u00ecm deal mi\u1ec5n ph\u00ed/gi\u1ea3m gi\u00e1 li\u00ean quan AI & LLM trong ng\u00e0y \u2014 bao g\u1ed3m c\u1ea3 AI coding tools, IDE, v\u00e0 developer platforms.\\nY\u00eau c\u1ea7u:\\n1) Ph\u1ea1m vi: LLM API providers, AI coding assistants (Kiro, Cursor, Windsurf, Copilot, Cody, v.v.), AI platforms, developer tools c\u00f3 t\u00edch h\u1ee3p AI.\\n2) T\u1eadp trung ngu\u1ed3n ch\u00ednh th\u1ed1ng: trang pricing/changelog/blog ch\u00ednh th\u1ee9c, \u0111\u1ed1i t\u00e1c thanh to\u00e1n, coupon c\u00f3 uy t\u00edn.\\n3) Ch\u1ec9 li\u1ec7t k\u00ea deal c\u00f2n hi\u1ec7u l\u1ef1c ho\u1eb7c ch\u01b0a th\u1ea5y th\u00f4ng tin h\u1ebft h\u1ea1n.\\n4) M\u1ed7i deal g\u1ed3m: t\u00ean nh\u00e0 cung c\u1ea5p, \u01b0u \u0111\u00e3i, \u0111i\u1ec1u ki\u1ec7n \u00e1p d\u1ee5ng, h\u1ea1n ch\u00f3t (n\u1ebfu c\u00f3), link ngu\u1ed3n tr\u1ef1c ti\u1ebfp.\\n5) \u0110\u00e1nh d\u1ea5u m\u1ee9c \u0111\u1ed9 tin c\u1eady: Cao (ngu\u1ed3n ch\u00ednh th\u1ee9c) / Trung b\u00ecnh (ngu\u1ed3n t\u1ed5ng h\u1ee3p uy t\u00edn).\\n6) N\u1ebfu kh\u00f4ng c\u00f3 deal m\u1edbi trong ng\u00e0y, tr\u1ea3: \\\"H\u00f4m nay ch\u01b0a ghi nh\u1eadn deal AI/LLM m\u1edbi \u0111\u00e1ng tin c\u1eady\\\" v\u00e0 th\u00eam t\u1ed1i \u0111a 3 ch\u01b0\u01a1ng tr\u00ecnh c\u00f2n hi\u1ec7u l\u1ef1c g\u1ea7n nh\u1ea5t (n\u1ebfu x\u00e1c minh \u0111\u01b0\u1ee3c).\\n7) Tr\u00ecnh b\u00e0y d\u1ea1ng bullet, ng\u1eafn g\u1ecdn, d\u1ec5 \u0111\u1ecdc tr\u00ean Telegram.\\nCurrent time: Tuesday, June 9th, 2026 - 8:15 PM (Asia/Saigon)\\nReference UTC: 2026-06-09 13:15 UTC\\n\\nUse the message tool if you need to notify the user directly for the current chat. If you do not send directly, your final plain-text reply will be delivered automatically.\"}]}],\"model\":\"gpt-5.5\",\"stream\":true,\"max_completion_tokens\":4096,\"tools\":[{\"type\":\"function\",\"function\":{\"name\":\"agents_list\",\"description\":\"List agent ids allowed for `sessions_spawn runtime=\\\"subagent\\\"`.\",\"parameters\":{\"type\":\"object\",\"properties\":{}}}},{\"type\":\"function\",\"function\":{\"name\":\"browser\",\"description\":\"Control the browser via OpenClaw's browser control server (status/start/stop/profiles/tabs/open/snapshot/screenshot/actions). Browser choice: omit profile by default for the isolated OpenClaw-managed browser (`openclaw`). For the logged-in user browser, use profile=\\\"user\\\". A supported Chromium-based browser (v144+) must be running on the selected host or browser node. Use only when existing logins/cookies matter and the user is present. For profile=\\\"user\\\" or other existing-session profiles, omit timeoutMs on act:type, evaluate, hover, scrollIntoView, drag, select, and fill; that driver rejects per-call timeout overrides for those actions. When a node-hosted browser proxy is available, the tool may auto-route to it. Pin a node with node=<id|name> or target=\\\"node\\\". When using refs from snapshot (e.g. e12), keep the same tab: prefer passing targetId from the snapshot response into subsequent actions (act/click/type/etc). For tab operations, targetId also accepts tabId handles (t1) and labels from action=tabs. For multi-step browser work, login checks, stale refs, duplicate tabs, or Google Meet flows, use the bundled browser-automation skill when it is available. For stable, self-resolving refs across calls, use snapshot with refs=\\\"aria\\\" (Playwright aria-ref ids). Default refs=\\\"role\\\" are role+name-based. Use snapshot+act for UI automation. Avoid act:wait by default; use only in exceptional cases when no reliable UI state exists. target selects browser location (sandbox|host|node). Default: host. Host target allowed.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"doctor\",\"status\",\"start\",\"stop\",\"profiles\",\"tabs\",\"open\",\"focus\",\"close\",\"snapshot\",\"screenshot\",\"navigate\",\"console\",\"pdf\",\"upload\",\"dialog\",\"act\"]},\"target\":{\"type\":\"string\",\"enum\":[\"sandbox\",\"host\",\"node\"]},\"node\":{\"type\":\"string\"},\"profile\":{\"type\":\"string\"},\"targetUrl\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"targetId\":{\"type\":\"string\"},\"label\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\"},\"maxChars\":{\"type\":\"number\"},\"mode\":{\"type\":\"string\",\"enum\":[\"efficient\"]},\"snapshotFormat\":{\"type\":\"string\",\"enum\":[\"aria\",\"ai\"]},\"refs\":{\"type\":\"string\",\"enum\":[\"role\",\"aria\"]},\"interactive\":{\"type\":\"boolean\"},\"compact\":{\"type\":\"boolean\"},\"depth\":{\"type\":\"number\"},\"selector\":{\"type\":\"string\"},\"frame\":{\"type\":\"string\"},\"labels\":{\"type\":\"boolean\"},\"urls\":{\"type\":\"boolean\"},\"fullPage\":{\"type\":\"boolean\"},\"ref\":{\"type\":\"string\"},\"element\":{\"type\":\"string\"},\"type\":{\"type\":\"string\",\"enum\":[\"png\",\"jpeg\"]},\"level\":{\"type\":\"string\"},\"paths\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"inputRef\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"dialogId\":{\"type\":\"string\"},\"accept\":{\"type\":\"boolean\"},\"promptText\":{\"type\":\"string\"},\"kind\":{\"type\":\"string\",\"enum\":[\"click\",\"clickCoords\",\"type\",\"press\",\"hover\",\"drag\",\"select\",\"fill\",\"resize\",\"wait\",\"evaluate\",\"close\"]},\"doubleClick\":{\"type\":\"boolean\"},\"button\":{\"type\":\"string\"},\"modifiers\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"x\":{\"type\":\"number\"},\"y\":{\"type\":\"number\"},\"text\":{\"type\":\"string\"},\"submit\":{\"type\":\"boolean\"},\"slowly\":{\"type\":\"boolean\"},\"key\":{\"type\":\"string\"},\"delayMs\":{\"type\":\"number\"},\"startRef\":{\"type\":\"string\"},\"endRef\":{\"type\":\"string\"},\"values\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"fields\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{},\"additionalProperties\":true}},\"width\":{\"type\":\"number\"},\"height\":{\"type\":\"number\"},\"timeMs\":{\"type\":\"number\"},\"textGone\":{\"type\":\"string\"},\"loadState\":{\"type\":\"string\"},\"fn\":{\"type\":\"string\"},\"request\":{\"type\":\"object\",\"required\":[\"kind\"],\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"click\",\"clickCoords\",\"type\",\"press\",\"hover\",\"drag\",\"select\",\"fill\",\"resize\",\"wait\",\"evaluate\",\"close\"]},\"targetId\":{\"type\":\"string\"},\"ref\":{\"type\":\"string\"},\"doubleClick\":{\"type\":\"boolean\"},\"button\":{\"type\":\"string\"},\"modifiers\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"x\":{\"type\":\"number\"},\"y\":{\"type\":\"number\"},\"text\":{\"type\":\"string\"},\"submit\":{\"type\":\"boolean\"},\"slowly\":{\"type\":\"boolean\"},\"key\":{\"type\":\"string\"},\"delayMs\":{\"type\":\"number\"},\"startRef\":{\"type\":\"string\"},\"endRef\":{\"type\":\"string\"},\"values\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"fields\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{},\"additionalProperties\":true}},\"width\":{\"type\":\"number\"},\"height\":{\"type\":\"number\"},\"timeMs\":{\"type\":\"number\"},\"selector\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"loadState\":{\"type\":\"string\"},\"textGone\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"fn\":{\"type\":\"string\"}}}}}}},{\"type\":\"function\",\"function\":{\"name\":\"canvas\",\"description\":\"Control node canvases (present/hide/navigate/eval/snapshot/A2UI). Use snapshot to capture the rendered UI.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"present\",\"hide\",\"navigate\",\"eval\",\"snapshot\",\"a2ui_push\",\"a2ui_reset\"]},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"node\":{\"type\":\"string\"},\"target\":{\"type\":\"string\"},\"x\":{\"type\":\"number\"},\"y\":{\"type\":\"number\"},\"width\":{\"type\":\"number\"},\"height\":{\"type\":\"number\"},\"url\":{\"type\":\"string\"},\"javaScript\":{\"type\":\"string\"},\"outputFormat\":{\"type\":\"string\",\"enum\":[\"png\",\"jpg\",\"jpeg\"]},\"maxWidth\":{\"type\":\"number\"},\"quality\":{\"type\":\"number\"},\"delayMs\":{\"type\":\"number\"},\"jsonl\":{\"type\":\"string\"},\"jsonlPath\":{\"type\":\"string\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"cron\",\"description\":\"Manage Gateway cron jobs and wake events: reminders, check-back-later, delayed follow-ups, recurring work. Do not emulate scheduling with exec sleep/process polling.\\n\\nMain cron => system events for heartbeat. Isolated cron => background task in `openclaw tasks`.\\n\\nACTIONS:\\n- status: scheduler status\\n- list: jobs; includeDisabled true includes disabled; agentId filter auto-filled from session\\n- get: one job; needs jobId\\n- add: create job; needs job object\\n- update: patch job; needs jobId + patch\\n- remove: delete job; needs jobId\\n- run: trigger now; needs jobId\\n- runs: run history; needs jobId\\n- wake: send wake event; needs text, optional mode\\n\\nJOB SCHEMA (for add action):\\n{\\n  \\\"name\\\": \\\"string\\\",\\n  \\\"schedule\\\": { ... },      // required\\n  \\\"payload\\\": { ... },       // required\\n  \\\"delivery\\\": { ... },      // optional announce for isolated/current/session, webhook for any target\\n  \\\"sessionTarget\\\": \\\"main\\\" | \\\"isolated\\\" | \\\"current\\\" | \\\"session:<id>\\\",\\n  \\\"enabled\\\": true | false   // default true\\n}\\n\\nSESSION TARGET OPTIONS:\\n- \\\"main\\\": main session; requires payload.kind=\\\"systemEvent\\\"\\n- \\\"isolated\\\": ephemeral isolated session; requires payload.kind=\\\"agentTurn\\\"\\n- \\\"current\\\": bind current session at creation\\n- \\\"session:<id>\\\": persistent named session\\n\\nDEFAULTS:\\n- payload.kind=\\\"systemEvent\\\" \u2192 defaults to \\\"main\\\"\\n- payload.kind=\\\"agentTurn\\\" \u2192 defaults to \\\"isolated\\\"\\nCurrent binding needs sessionTarget=\\\"current\\\".\\n\\nSCHEDULE TYPES (schedule.kind):\\n- \\\"at\\\": one-shot absolute time\\n  { \\\"kind\\\": \\\"at\\\", \\\"at\\\": \\\"<ISO-8601 timestamp>\\\" }\\n- \\\"every\\\": recurring interval\\n  { \\\"kind\\\": \\\"every\\\", \\\"everyMs\\\": <ms>, \\\"anchorMs\\\": <optional-ms> }\\n- \\\"cron\\\": expr in supplied timezone, or Gateway host local timezone when tz omitted\\n  { \\\"kind\\\": \\\"cron\\\", \\\"expr\\\": \\\"<cron-expression>\\\", \\\"tz\\\": \\\"<optional-IANA-timezone>\\\" }\\n  Write expr in local wall-clock time; do not convert the requested local time to UTC first.\\n  tz omitted => Gateway host local timezone, not UTC.\\n  Example 6pm Shanghai daily: { \\\"kind\\\": \\\"cron\\\", \\\"expr\\\": \\\"0 18 * * *\\\", \\\"tz\\\": \\\"Asia/Shanghai\\\" }\\n\\nFor \\\"at\\\", ISO timestamps without timezone are UTC.\\n\\nPAYLOAD TYPES (payload.kind):\\n- \\\"systemEvent\\\": inject text as system event\\n  { \\\"kind\\\": \\\"systemEvent\\\", \\\"text\\\": \\\"<message>\\\" }\\n- \\\"agentTurn\\\": run agent with prompt; isolated/current/session only\\n  { \\\"kind\\\": \\\"agentTurn\\\", \\\"message\\\": \\\"<prompt>\\\", \\\"model\\\": \\\"<optional>\\\", \\\"thinking\\\": \\\"<optional>\\\", \\\"timeoutSeconds\\\": <optional, 0=no timeout> }\\n\\nDELIVERY (top-level):\\n  { \\\"mode\\\": \\\"none|announce|webhook\\\", \\\"channel\\\": \\\"<optional>\\\", \\\"to\\\": \\\"<optional>\\\", \\\"threadId\\\": \\\"<optional>\\\", \\\"bestEffort\\\": <optional-bool> }\\n  - isolated agentTurn default when omitted: \\\"announce\\\"\\n  - announce: send to chat channel; isolated/current/session only; optional channel/to\\n  - threadId: chat thread/topic id\\n  - webhook: POST finished-run event to delivery.to URL\\n  - Specific chat/recipient: set announce delivery.channel/to; do not call messaging tools inside run.\\n\\nCRITICAL CONSTRAINTS:\\n- sessionTarget=\\\"main\\\" REQUIRES payload.kind=\\\"systemEvent\\\"\\n- sessionTarget=\\\"isolated\\\" | \\\"current\\\" | \\\"session:xxx\\\" REQUIRES payload.kind=\\\"agentTurn\\\"\\n- Webhook: delivery.mode=\\\"webhook\\\" and delivery.to URL.\\nDefault: prefer isolated agentTurn jobs unless the user explicitly wants current-session binding.\\n\\nRESTRICTED CRON RUNS:\\n- Some isolated cron runs get narrow self-cleanup grant: status/list self-only, get/runs current job only, mutation only remove current job.\\n\\nWAKE MODES (for wake action):\\n- \\\"next-heartbeat\\\" default: wake next heartbeat\\n- \\\"now\\\": wake immediately\\n\\nUse jobId canonical; id accepted compat. contextMessages (0-10) adds previous messages as job context.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"status\",\"list\",\"get\",\"add\",\"update\",\"remove\",\"run\",\"runs\",\"wake\"]},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"includeDisabled\":{\"type\":\"boolean\"},\"job\":{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\",\"description\":\"Job name\"},\"schedule\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"at\",\"every\",\"cron\"],\"description\":\"Schedule kind\"},\"at\":{\"type\":\"string\",\"description\":\"ISO-8601 time (kind=at)\"},\"everyMs\":{\"type\":\"number\",\"description\":\"Interval ms (kind=every)\"},\"anchorMs\":{\"type\":\"number\",\"description\":\"Start anchor ms (kind=every)\"},\"expr\":{\"type\":\"string\",\"description\":\"Cron expr in tz wall-clock time; do not convert to UTC. Omitted tz => Gateway host local timezone. Example 6pm Shanghai daily: expr \\\"0 18 * * *\\\", tz \\\"Asia/Shanghai\\\".\"},\"tz\":{\"type\":\"string\",\"description\":\"IANA timezone for cron wall-clock fields, e.g. \\\"Asia/Shanghai\\\"; omitted => Gateway host local timezone.\"},\"staggerMs\":{\"type\":\"number\",\"description\":\"Jitter ms (kind=cron)\"}},\"additionalProperties\":true},\"sessionTarget\":{\"type\":\"string\",\"description\":\"main | isolated | current | session:<id>\"},\"wakeMode\":{\"type\":\"string\",\"enum\":[\"now\",\"next-heartbeat\"],\"description\":\"Wake timing\"},\"payload\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"systemEvent\",\"agentTurn\"],\"description\":\"Payload kind\"},\"text\":{\"type\":\"string\",\"description\":\"systemEvent text\"},\"message\":{\"type\":\"string\",\"description\":\"agentTurn prompt\"},\"model\":{\"type\":\"string\",\"description\":\"Model override\"},\"thinking\":{\"type\":\"string\",\"description\":\"Thinking override\"},\"timeoutSeconds\":{\"type\":\"number\"},\"lightContext\":{\"type\":\"boolean\"},\"allowUnsafeExternalContent\":{\"type\":\"boolean\"},\"fallbacks\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Fallback models\"},\"toolsAllow\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Allowed tools\"}},\"additionalProperties\":true},\"delivery\":{\"type\":\"object\",\"properties\":{\"mode\":{\"type\":\"string\",\"enum\":[\"none\",\"announce\",\"webhook\"],\"description\":\"Delivery mode\"},\"channel\":{\"type\":\"string\",\"description\":\"Delivery channel\"},\"to\":{\"type\":\"string\",\"description\":\"Delivery target\"},\"threadId\":{\"anyOf\":[{\"type\":\"string\"},{\"type\":\"number\"}],\"description\":\"Thread/topic id\"},\"bestEffort\":{\"type\":\"boolean\"},\"accountId\":{\"type\":\"string\",\"description\":\"Delivery account\"},\"failureDestination\":{\"type\":\"object\",\"properties\":{\"channel\":{\"type\":\"string\"},\"to\":{\"type\":\"string\"},\"accountId\":{\"type\":\"string\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]}},\"additionalProperties\":true}},\"additionalProperties\":true},\"agentId\":{\"type\":\"string\",\"description\":\"Agent id, or null to keep it unset\"},\"description\":{\"type\":\"string\",\"description\":\"Human description\"},\"enabled\":{\"type\":\"boolean\"},\"deleteAfterRun\":{\"type\":\"boolean\",\"description\":\"Delete after first run\"},\"sessionKey\":{\"type\":\"string\",\"description\":\"Explicit session key, or null to clear it\"},\"failureAlert\":{\"type\":\"object\",\"properties\":{\"after\":{\"type\":\"number\",\"description\":\"Failures before alert\"},\"channel\":{\"type\":\"string\",\"description\":\"Alert channel\"},\"to\":{\"type\":\"string\",\"description\":\"Alert target\"},\"cooldownMs\":{\"type\":\"number\",\"description\":\"Alert cooldown ms\"},\"includeSkipped\":{\"type\":\"boolean\",\"description\":\"Skipped runs count toward alert\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]},\"accountId\":{\"type\":\"string\"}},\"additionalProperties\":true,\"description\":\"Failure alert object; false disables alerts\"}},\"additionalProperties\":true},\"jobId\":{\"type\":\"string\"},\"id\":{\"type\":\"string\"},\"patch\":{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\",\"description\":\"Job name\"},\"schedule\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"at\",\"every\",\"cron\"],\"description\":\"Schedule kind\"},\"at\":{\"type\":\"string\",\"description\":\"ISO-8601 time (kind=at)\"},\"everyMs\":{\"type\":\"number\",\"description\":\"Interval ms (kind=every)\"},\"anchorMs\":{\"type\":\"number\",\"description\":\"Start anchor ms (kind=every)\"},\"expr\":{\"type\":\"string\",\"description\":\"Cron expr in tz wall-clock time; do not convert to UTC. Omitted tz => Gateway host local timezone. Example 6pm Shanghai daily: expr \\\"0 18 * * *\\\", tz \\\"Asia/Shanghai\\\".\"},\"tz\":{\"type\":\"string\",\"description\":\"IANA timezone for cron wall-clock fields, e.g. \\\"Asia/Shanghai\\\"; omitted => Gateway host local timezone.\"},\"staggerMs\":{\"type\":\"number\",\"description\":\"Jitter ms (kind=cron)\"}},\"additionalProperties\":true},\"sessionTarget\":{\"type\":\"string\",\"description\":\"Session target\"},\"wakeMode\":{\"type\":\"string\",\"enum\":[\"now\",\"next-heartbeat\"]},\"payload\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"systemEvent\",\"agentTurn\"],\"description\":\"Payload kind\"},\"text\":{\"type\":\"string\",\"description\":\"systemEvent text\"},\"message\":{\"type\":\"string\",\"description\":\"agentTurn prompt\"},\"model\":{\"type\":\"string\",\"description\":\"Model override\"},\"thinking\":{\"type\":\"string\",\"description\":\"Thinking override\"},\"timeoutSeconds\":{\"type\":\"number\"},\"lightContext\":{\"type\":\"boolean\"},\"allowUnsafeExternalContent\":{\"type\":\"boolean\"},\"fallbacks\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Fallback models\"},\"toolsAllow\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Allowed tool ids, or null to clear\"}},\"additionalProperties\":true},\"delivery\":{\"type\":\"object\",\"properties\":{\"mode\":{\"type\":\"string\",\"enum\":[\"none\",\"announce\",\"webhook\"],\"description\":\"Delivery mode\"},\"channel\":{\"type\":\"string\",\"description\":\"Delivery channel\"},\"to\":{\"type\":\"string\",\"description\":\"Delivery target\"},\"threadId\":{\"anyOf\":[{\"type\":\"string\"},{\"type\":\"number\"}],\"description\":\"Thread/topic id\"},\"bestEffort\":{\"type\":\"boolean\"},\"accountId\":{\"type\":\"string\",\"description\":\"Delivery account\"},\"failureDestination\":{\"type\":\"object\",\"properties\":{\"channel\":{\"type\":\"string\"},\"to\":{\"type\":\"string\"},\"accountId\":{\"type\":\"string\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]}},\"additionalProperties\":true}},\"additionalProperties\":true},\"description\":{\"type\":\"string\"},\"enabled\":{\"type\":\"boolean\"},\"deleteAfterRun\":{\"type\":\"boolean\"},\"agentId\":{\"type\":\"string\",\"description\":\"Agent id, or null to clear it\"},\"sessionKey\":{\"type\":\"string\",\"description\":\"Explicit session key, or null to clear it\"},\"failureAlert\":{\"type\":\"object\",\"properties\":{\"after\":{\"type\":\"number\",\"description\":\"Failures before alert\"},\"channel\":{\"type\":\"string\",\"description\":\"Alert channel\"},\"to\":{\"type\":\"string\",\"description\":\"Alert target\"},\"cooldownMs\":{\"type\":\"number\",\"description\":\"Alert cooldown ms\"},\"includeSkipped\":{\"type\":\"boolean\",\"description\":\"Skipped runs count toward alert\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]},\"accountId\":{\"type\":\"string\"}},\"additionalProperties\":true,\"description\":\"Failure alert object; false disables alerts\"}},\"additionalProperties\":true},\"text\":{\"type\":\"string\"},\"mode\":{\"type\":\"string\",\"enum\":[\"now\",\"next-heartbeat\"]},\"runMode\":{\"type\":\"string\",\"enum\":[\"due\",\"force\"]},\"contextMessages\":{\"type\":\"number\",\"minimum\":0,\"maximum\":10},\"agentId\":{\"type\":\"string\",\"description\":\"List filter: agent id\"}},\"additionalProperties\":true}}},{\"type\":\"function\",\"function\":{\"name\":\"dir_fetch\",\"description\":\"Retrieve a directory tree from a paired node as a gzipped tarball, unpack it on the gateway, and return a manifest of saved paths. Use to pull source trees, asset folders, or log directories in a single round-trip. The unpacked files live on the GATEWAY (not your local machine); pass localPath into other tools or use file_fetch on individual entries to ship them elsewhere. Rejects trees larger than 16 MB compressed. Requires operator opt-in: gateway.nodes.allowCommands must include 'dir.fetch' AND plugins.entries.file-transfer.config.nodes.<node>.allowReadPaths must match the directory path.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id, name, or IP. Resolves the same way as the nodes tool.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path to the directory on the node to fetch. Canonicalized server-side.\"},\"maxBytes\":{\"type\":\"number\",\"description\":\"Max gzipped tarball bytes to fetch. Default 8 MB, hard ceiling 16 MB (single round-trip).\"},\"includeDotfiles\":{\"type\":\"boolean\",\"description\":\"Reserved for v2; currently always includes dotfiles (v1 quirk in BSD tar).\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"dir_list\",\"description\":\"Retrieve a structured directory listing from a paired node. Returns file and subdirectory metadata (name, path, size, mimeType, isDir, mtime) without transferring file content. Use this to discover what files exist before fetching them with file_fetch. Pagination is offset-based; pass nextPageToken from the previous result. Requires operator opt-in: gateway.nodes.allowCommands must include 'dir.list' AND plugins.entries.file-transfer.config.nodes.<node>.allowReadPaths must match the directory path. Without policy configured, every call is denied.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id, name, or IP. Resolves the same way as the nodes tool.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path to the directory on the node. Canonicalized server-side.\"},\"pageToken\":{\"type\":\"string\",\"description\":\"Pagination token from a previous dir_list call. Omit to start from the beginning.\"},\"maxEntries\":{\"type\":\"number\",\"description\":\"Max entries per page. Default 200, hard ceiling 5000.\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"edit\",\"description\":\"Edit a single file using exact text replacement. Every edits[].oldText must match a unique, non-overlapping region of the original file. If two changes affect the same block or nearby lines, merge them into one edit instead of emitting overlapping edits. Do not include large unchanged regions just to connect distant changes.\",\"parameters\":{\"type\":\"object\",\"required\":[\"path\",\"edits\"],\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Path to the file to edit (relative or absolute)\"},\"edits\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"oldText\",\"newText\"],\"properties\":{\"oldText\":{\"type\":\"string\",\"description\":\"Exact text for one targeted replacement. It must be unique in the original file and must not overlap with any other edits[].oldText in the same call.\"},\"newText\":{\"type\":\"string\",\"description\":\"Replacement text for this targeted edit.\"}},\"additionalProperties\":false},\"description\":\"One or more targeted replacements. Each edit is matched against the original file, not incrementally. Do not include overlapping or nested edits. If two changes touch the same block or nearby lines, merge them into one edit instead.\"}},\"additionalProperties\":false}}},{\"type\":\"function\",\"function\":{\"name\":\"exec\",\"description\":\"Execute shell commands with background continuation for work that starts now. Use yieldMs/background to continue later via process tool. For long-running work started now, rely on automatic completion wake when it is enabled and the command emits output or fails; otherwise use process to confirm completion. Use process whenever you need logs, status, input, or intervention. Do not use exec sleep or delay loops for reminders or deferred follow-ups; use cron instead. Use pty=true for TTY-required commands (terminal UIs, coding agents).\",\"parameters\":{\"type\":\"object\",\"required\":[\"command\"],\"properties\":{\"command\":{\"type\":\"string\",\"description\":\"Shell command to execute\"},\"workdir\":{\"type\":\"string\",\"description\":\"Working directory (defaults to cwd)\"},\"env\":{\"type\":\"object\",\"patternProperties\":{\"^.*$\":{\"type\":\"string\"}}},\"yieldMs\":{\"type\":\"number\",\"description\":\"Milliseconds to wait before backgrounding (default 10000)\"},\"background\":{\"type\":\"boolean\",\"description\":\"Run in background immediately\"},\"timeout\":{\"type\":\"number\",\"description\":\"Timeout in seconds (optional, kills process on expiry)\"},\"pty\":{\"type\":\"boolean\",\"description\":\"Run in a pseudo-terminal (PTY) when available (TTY-required CLIs, coding agents)\"},\"elevated\":{\"type\":\"boolean\",\"description\":\"Run on the host with elevated permissions (if allowed)\"},\"host\":{\"type\":\"string\",\"enum\":[\"auto\",\"sandbox\",\"gateway\",\"node\"],\"description\":\"Exec host/target (auto|sandbox|gateway|node).\"},\"security\":{\"type\":\"string\",\"description\":\"Ignored for normal calls; exec security is set by tools.exec.security and host approvals.\"},\"ask\":{\"type\":\"string\",\"description\":\"Exec ask mode (off|on-miss|always).\"},\"node\":{\"type\":\"string\",\"description\":\"Node id/name for host=node.\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"file_fetch\",\"description\":\"Retrieve a file from a paired node by absolute path. Returns image content blocks for image MIME types, inlines small text files (\u22648 KB) as text content, and saves everything else under the gateway media store with a path you can pass to file_write or other tools. Use this for screenshots, photos, receipts, logs, source files. Pair with file_write to copy a file from one node to another (no exec/cp shell-out needed). Requires operator opt-in: gateway.nodes.allowCommands must include 'file.fetch' AND plugins.entries.file-transfer.config.nodes.<node>.allowReadPaths must match the path. Without policy configured, every call is denied.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id, name, or IP. Resolves the same way as the nodes tool.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path to the file on the node. Canonicalized server-side.\"},\"maxBytes\":{\"type\":\"number\",\"description\":\"Max bytes to fetch. Default 8 MB, hard ceiling 16 MB (single round-trip).\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"file_write\",\"description\":\"Write file bytes to a paired node by absolute path. Atomic write (temp + rename). Refuses to overwrite by default \u2014 pass overwrite=true to replace. Refuses to write through symlink targets unless policy explicitly allows following symlinks. Pair with file_fetch by passing its mediaId as sourceMediaId for binary copy. Requires operator opt-in: gateway.nodes.allowCommands must include 'file.write' AND plugins.entries.file-transfer.config.nodes.<node>.allowWritePaths must match the destination path. Without policy configured, every call is denied.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id or display name to write the file on.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path on the node to write. Canonicalized server-side.\"},\"contentBase64\":{\"type\":\"string\",\"description\":\"Base64-encoded bytes to write. Maximum 16 MB after decode.\"},\"sourceMediaId\":{\"type\":\"string\",\"description\":\"Media id returned by file_fetch. Preferred for binary copies because bytes stay in the gateway media store.\"},\"mimeType\":{\"type\":\"string\",\"description\":\"Content type hint. Not validated against the content.\"},\"overwrite\":{\"type\":\"boolean\",\"description\":\"Allow overwriting an existing file. Default false.\",\"default\":false},\"createParents\":{\"type\":\"boolean\",\"description\":\"Create missing parent directories (mkdir -p). Default false.\",\"default\":false}}}}},{\"type\":\"function\",\"function\":{\"name\":\"gateway\",\"description\":\"Gateway restart/config/update. Before config edits, use config.schema.lookup with targeted dot path. Prefer config.patch for partial merge; config.apply only full replace. Writes hot-reload or restart as needed. Always pass human `note` for post-restart delivery. If still owe the user a reply, pass one-shot `continuationMessage`; do not write restart sentinel files directly.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"restart\",\"config.get\",\"config.schema.lookup\",\"config.apply\",\"config.patch\",\"update.run\"]},\"delayMs\":{\"type\":\"number\"},\"reason\":{\"type\":\"string\"},\"continuationMessage\":{\"type\":\"string\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"path\":{\"type\":\"string\"},\"raw\":{\"type\":\"string\"},\"baseHash\":{\"type\":\"string\"},\"sessionKey\":{\"type\":\"string\"},\"note\":{\"type\":\"string\"},\"restartDelayMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"image\",\"description\":\"Analyze images with available vision model. Use image for one path/URL, images for max 20. Prompt says what to inspect.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"prompt\":{\"type\":\"string\"},\"image\":{\"type\":\"string\",\"description\":\"One image path/URL.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Image paths/URLs; maxImages default 20.\"},\"model\":{\"type\":\"string\"},\"maxBytesMb\":{\"type\":\"number\"},\"maxImages\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"image_generate\",\"description\":\"Create/edit images. Session chats: background task; do not call image_generate again for same request; wait completion, then send attachments via message tool. Transparent: outputFormat=\\\"png\\\" or \\\"webp\\\" + background=\\\"transparent\\\"; OpenAI also supports openai.background and routes default model to gpt-image-1.5. Use action=\\\"list\\\" for providers/models/readiness/auth, \\\"status\\\" for active task.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"\\\"generate\\\" default, \\\"status\\\" active task, \\\"list\\\" providers/models.\"},\"prompt\":{\"type\":\"string\",\"description\":\"Image prompt.\"},\"image\":{\"type\":\"string\",\"description\":\"Reference image path/URL for edit.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference images for edit; max 5.\"},\"model\":{\"type\":\"string\",\"description\":\"Provider/model override, e.g. openai/gpt-image-2; transparent OpenAI: openai/gpt-image-1.5.\"},\"filename\":{\"type\":\"string\",\"description\":\"Output filename hint; basename preserved in managed media dir.\"},\"size\":{\"type\":\"string\",\"description\":\"Size hint: 1024x1024, 1536x1024, 1024x1536, 2048x2048, 3840x2160.\"},\"aspectRatio\":{\"type\":\"string\",\"description\":\"Aspect ratio: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.\"},\"resolution\":{\"type\":\"string\",\"description\":\"Resolution: 1K, 2K, 4K; useful for Google.\"},\"quality\":{\"type\":\"string\",\"enum\":[\"low\",\"medium\",\"high\",\"auto\"],\"description\":\"Quality: low, medium, high, auto.\"},\"outputFormat\":{\"type\":\"string\",\"enum\":[\"png\",\"jpeg\",\"webp\"],\"description\":\"Output format: png, jpeg, webp.\"},\"background\":{\"type\":\"string\",\"enum\":[\"transparent\",\"opaque\",\"auto\"],\"description\":\"Background: transparent, opaque, auto. Transparent needs png/webp output.\"},\"openai\":{\"type\":\"object\",\"properties\":{\"background\":{\"type\":\"string\",\"enum\":[\"transparent\",\"opaque\",\"auto\"],\"description\":\"OpenAI background: transparent, opaque, auto. Transparent needs png/webp; default model routes to gpt-image-1.5.\"},\"moderation\":{\"type\":\"string\",\"enum\":[\"low\",\"auto\"],\"description\":\"OpenAI moderation: low, auto.\"},\"outputCompression\":{\"type\":\"number\",\"description\":\"OpenAI jpeg/webp compression 0-100.\",\"minimum\":0,\"maximum\":100},\"user\":{\"type\":\"string\",\"description\":\"OpenAI stable end-user id.\"}}},\"count\":{\"type\":\"number\",\"description\":\"Image count 1-4.\",\"minimum\":1,\"maximum\":4},\"timeoutMs\":{\"type\":\"number\",\"description\":\"Provider timeout ms.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"memory_get\",\"description\":\"Safe exact excerpt read from MEMORY.md or memory/*.md. Defaults to a bounded excerpt when lines are omitted, includes truncation/continuation info when more content exists, and `corpus=wiki` reads from registered compiled-wiki supplements.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"path\":{\"type\":\"string\"},\"from\":{\"type\":\"number\"},\"lines\":{\"type\":\"number\"},\"corpus\":{\"type\":\"string\",\"enum\":[\"memory\",\"wiki\",\"all\"]}},\"required\":[\"path\"],\"additionalProperties\":false}}},{\"type\":\"function\",\"function\":{\"name\":\"memory_search\",\"description\":\"Mandatory recall step: semantically search MEMORY.md + memory/*.md (and optional session transcripts) before answering questions about prior work, decisions, dates, people, preferences, or todos. Optional `corpus=wiki` or `corpus=all` also searches registered compiled-wiki supplements. `corpus=memory` restricts hits to indexed memory files (excludes session transcript chunks from ranking). `corpus=sessions` restricts hits to indexed session transcripts (same visibility rules as session history tools). If response has disabled=true, memory retrieval is unavailable and should be surfaced to the user.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"},\"maxResults\":{\"type\":\"number\"},\"minScore\":{\"type\":\"number\"},\"corpus\":{\"type\":\"string\",\"enum\":[\"memory\",\"wiki\",\"all\",\"sessions\"]}},\"required\":[\"query\"],\"additionalProperties\":false}}},{\"type\":\"function\",\"function\":{\"name\":\"message\",\"description\":\"Send/delete/manage channel messages. Supports actions: delete, edit, poll, react, send, topic-create, topic-edit. This turn: use action=\\\"send\\\" with message for visible replies to the current source conversation. target defaults to the current source conversation; omit unless sending elsewhere. Normal final answers stay private.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"send\",\"poll\",\"react\",\"delete\",\"edit\",\"topic-create\",\"topic-edit\"]},\"channel\":{\"type\":\"string\"},\"target\":{\"type\":\"string\",\"description\":\"Recipient/channel: E.164 for WhatsApp/Signal, Telegram chat id/@username, Discord/Slack/Mattermost <channelId|user:ID|channel:ID>, or iMessage handle/chat_id\"},\"targets\":{\"type\":\"array\",\"items\":{\"type\":\"string\",\"description\":\"Recipient/channel targets (same format as --target); accepts ids or names when the directory is available.\"}},\"accountId\":{\"type\":\"string\"},\"dryRun\":{\"type\":\"boolean\"},\"message\":{\"type\":\"string\"},\"effectId\":{\"type\":\"string\",\"description\":\"Effect id/name for sendWithEffect.\"},\"effect\":{\"type\":\"string\",\"description\":\"Alias for effectId.\"},\"media\":{\"type\":\"string\",\"description\":\"Media URL/path. data: use buffer.\"},\"filename\":{\"type\":\"string\"},\"buffer\":{\"type\":\"string\",\"description\":\"Base64 attachment payload; data URL ok.\"},\"contentType\":{\"type\":\"string\"},\"mimeType\":{\"type\":\"string\"},\"caption\":{\"type\":\"string\"},\"path\":{\"type\":\"string\"},\"filePath\":{\"type\":\"string\"},\"attachments\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{\"type\":{\"type\":\"string\",\"enum\":[\"image\",\"audio\",\"video\",\"file\"]},\"media\":{\"type\":\"string\"},\"mediaUrl\":{\"type\":\"string\"},\"path\":{\"type\":\"string\"},\"filePath\":{\"type\":\"string\"},\"fileUrl\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"name\":{\"type\":\"string\"},\"mimeType\":{\"type\":\"string\"}}},\"description\":\"Structured attachments; each needs media/mediaUrl/path/filePath/fileUrl/url.\"},\"replyTo\":{\"type\":\"string\"},\"threadId\":{\"type\":\"string\"},\"asVoice\":{\"type\":\"boolean\"},\"silent\":{\"type\":\"boolean\"},\"quoteText\":{\"type\":\"string\",\"description\":\"Telegram reply quote text.\"},\"bestEffort\":{\"type\":\"boolean\"},\"gifPlayback\":{\"type\":\"boolean\"},\"forceDocument\":{\"type\":\"boolean\",\"description\":\"Send image/GIF/video as document; avoids compression.\"},\"asDocument\":{\"type\":\"boolean\",\"description\":\"Alias for forceDocument.\"},\"presentation\":{\"type\":\"object\",\"required\":[\"blocks\"],\"properties\":{\"title\":{\"type\":\"string\"},\"tone\":{\"type\":\"string\",\"enum\":[\"info\",\"success\",\"warning\",\"danger\",\"neutral\"]},\"blocks\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"type\"],\"properties\":{\"type\":{\"type\":\"string\",\"enum\":[\"text\",\"context\",\"divider\",\"buttons\",\"select\"]},\"text\":{\"type\":\"string\"},\"buttons\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"label\"],\"properties\":{\"label\":{\"type\":\"string\"},\"value\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"webApp\":{\"type\":\"object\",\"required\":[\"url\"],\"properties\":{\"url\":{\"type\":\"string\"}}},\"web_app\":{\"type\":\"object\",\"required\":[\"url\"],\"properties\":{\"url\":{\"type\":\"string\"}}},\"disabled\":{\"type\":\"boolean\"},\"reusable\":{\"type\":\"boolean\"},\"style\":{\"type\":\"string\",\"enum\":[\"primary\",\"secondary\",\"success\",\"danger\"]}}}},\"placeholder\":{\"type\":\"string\"},\"options\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"label\",\"value\"],\"properties\":{\"label\":{\"type\":\"string\"},\"value\":{\"type\":\"string\"}}}}}}}},\"description\":\"Rich message payload: text/buttons/selects/context. Unsupported blocks degrade to text.\"},\"delivery\":{\"type\":\"object\",\"properties\":{\"pin\":{\"anyOf\":[{\"type\":\"boolean\"},{\"type\":\"object\",\"required\":[\"enabled\"],\"properties\":{\"enabled\":{\"type\":\"boolean\"},\"notify\":{\"type\":\"boolean\"},\"required\":{\"type\":\"boolean\"}}}]}},\"description\":\"Delivery prefs. pin requests pin when channel supports it.\"},\"messageId\":{\"type\":\"string\",\"description\":\"Target message id for read/react/edit/delete/pin/unpin. Reaction-like defaults current inbound id when available.\"},\"message_id\":{\"type\":\"string\",\"description\":\"snake_case alias of messageId; same defaults.\"},\"emoji\":{\"type\":\"string\"},\"remove\":{\"type\":\"boolean\"},\"trackToolCalls\":{\"type\":\"boolean\",\"description\":\"For current-message reaction, make reacted message the tool-progress reaction target.\"},\"track_tool_calls\":{\"type\":\"boolean\",\"description\":\"snake_case alias of trackToolCalls.\"},\"targetAuthor\":{\"type\":\"string\"},\"targetAuthorUuid\":{\"type\":\"string\"},\"groupId\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\"},\"pageSize\":{\"type\":\"number\"},\"pageToken\":{\"type\":\"string\"},\"before\":{\"type\":\"string\"},\"after\":{\"type\":\"string\"},\"around\":{\"type\":\"string\"},\"fromMe\":{\"type\":\"boolean\"},\"includeArchived\":{\"type\":\"boolean\"},\"pollId\":{\"type\":\"string\"},\"pollOptionId\":{\"type\":\"string\",\"description\":\"Poll answer id.\"},\"pollOptionIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\",\"description\":\"Poll answer ids for multiselect.\"}},\"pollOptionIndex\":{\"type\":\"number\",\"description\":\"1-based poll option number.\"},\"pollOptionIndexes\":{\"type\":\"array\",\"items\":{\"type\":\"number\",\"description\":\"1-based poll option numbers for multiselect.\"}},\"pollQuestion\":{\"type\":\"string\"},\"pollOption\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"pollDurationHours\":{\"type\":\"number\"},\"pollMulti\":{\"type\":\"boolean\"},\"channelId\":{\"type\":\"string\",\"description\":\"Channel id filter.\"},\"chatId\":{\"type\":\"string\",\"description\":\"Chat id for chat metadata.\"},\"channelIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\",\"description\":\"Channel id filter.\"}},\"memberId\":{\"type\":\"string\"},\"memberIdType\":{\"type\":\"string\"},\"guildId\":{\"type\":\"string\"},\"userId\":{\"type\":\"string\"},\"openId\":{\"type\":\"string\"},\"unionId\":{\"type\":\"string\"},\"authorId\":{\"type\":\"string\"},\"authorIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"roleId\":{\"type\":\"string\"},\"roleIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"participant\":{\"type\":\"string\"},\"includeMembers\":{\"type\":\"boolean\"},\"members\":{\"type\":\"boolean\"},\"scope\":{\"type\":\"string\"},\"kind\":{\"type\":\"string\"},\"fileId\":{\"type\":\"string\"},\"emojiName\":{\"type\":\"string\"},\"stickerId\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"stickerName\":{\"type\":\"string\"},\"stickerDesc\":{\"type\":\"string\"},\"stickerTags\":{\"type\":\"string\"},\"threadName\":{\"type\":\"string\"},\"autoArchiveMin\":{\"type\":\"number\"},\"appliedTags\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"query\":{\"type\":\"string\"},\"eventName\":{\"type\":\"string\"},\"eventType\":{\"type\":\"string\"},\"startTime\":{\"type\":\"string\"},\"endTime\":{\"type\":\"string\"},\"desc\":{\"type\":\"string\"},\"location\":{\"type\":\"string\"},\"image\":{\"type\":\"string\",\"description\":\"Event cover image URL/path.\"},\"durationMin\":{\"type\":\"number\"},\"until\":{\"type\":\"string\"},\"reason\":{\"type\":\"string\"},\"deleteDays\":{\"type\":\"number\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"name\":{\"type\":\"string\"},\"channelType\":{\"type\":\"number\",\"description\":\"Numeric channel type, e.g. Discord. Avoids JSON Schema `type` collision.\"},\"parentId\":{\"type\":\"string\"},\"topic\":{\"type\":\"string\"},\"position\":{\"type\":\"number\"},\"nsfw\":{\"type\":\"boolean\"},\"rateLimitPerUser\":{\"type\":\"number\"},\"categoryId\":{\"type\":\"string\"},\"clearParent\":{\"type\":\"boolean\",\"description\":\"Clear parent/category when supported.\"},\"activityType\":{\"type\":\"string\",\"description\":\"Activity type: playing, streaming, listening, watching, competing, custom.\"},\"activityName\":{\"type\":\"string\",\"description\":\"Activity name shown in sidebar; ignored for custom.\"},\"activityUrl\":{\"type\":\"string\",\"description\":\"Streaming URL; streaming type only.\"},\"activityState\":{\"type\":\"string\",\"description\":\"State text; custom type uses as status text.\"},\"status\":{\"type\":\"string\",\"description\":\"Bot status: online, dnd, idle, invisible.\"},\"pollDurationSeconds\":{\"type\":\"number\"},\"pollAnonymous\":{\"type\":\"boolean\"},\"pollPublic\":{\"type\":\"boolean\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"music_generate\",\"description\":\"Create audio/music for song, jingle, beat, loop, soundtrack, anthem, instrumental requests. If user asks make/generate/create song/music, call music_generate; do not just write lyrics unless lyrics/text only. Prompt gets style/genre/mood/tempo/instruments/purpose. lyrics only exact sung words. Session chats: background task; do not call again for same request; wait completion, send attachments via message tool. \\\"status\\\" checks active task.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"\\\"generate\\\" default, \\\"status\\\" active task, \\\"list\\\" providers/models.\"},\"prompt\":{\"type\":\"string\",\"description\":\"Music prompt: style, genre, mood, purpose.\"},\"lyrics\":{\"type\":\"string\",\"description\":\"Exact sung lyrics only when the user supplies lyrics or asks for vocal words. For song/style requests, use prompt instead.\"},\"instrumental\":{\"type\":\"boolean\",\"description\":\"Instrumental-only toggle.\"},\"image\":{\"type\":\"string\",\"description\":\"Reference image path/URL.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference images; max 10.\"},\"model\":{\"type\":\"string\",\"description\":\"Provider/model override, e.g. google/lyria-3-pro-preview.\"},\"durationSeconds\":{\"type\":\"number\",\"description\":\"Target seconds; provider may clamp.\",\"minimum\":1},\"format\":{\"type\":\"string\",\"description\":\"Output format: mp3, wav.\"},\"filename\":{\"type\":\"string\",\"description\":\"Output filename hint; basename preserved in managed media dir.\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"nodes\",\"description\":\"Discover/control paired nodes: status, describe, pairing, notify, camera/photos/screen/location/notifications/invoke. Use file_fetch for files.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"status\",\"describe\",\"pending\",\"approve\",\"reject\",\"notify\",\"camera_snap\",\"camera_list\",\"camera_clip\",\"photos_latest\",\"screen_record\",\"location_get\",\"notifications_list\",\"notifications_action\",\"device_status\",\"device_info\",\"device_permissions\",\"device_health\",\"invoke\"]},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"node\":{\"type\":\"string\"},\"requestId\":{\"type\":\"string\"},\"title\":{\"type\":\"string\"},\"body\":{\"type\":\"string\"},\"sound\":{\"type\":\"string\"},\"priority\":{\"type\":\"string\",\"enum\":[\"passive\",\"active\",\"timeSensitive\"]},\"delivery\":{\"type\":\"string\",\"enum\":[\"system\",\"overlay\",\"auto\"]},\"facing\":{\"type\":\"string\",\"enum\":[\"front\",\"back\",\"both\"],\"description\":\"camera_snap: front/back/both; camera_clip: front/back only.\"},\"maxWidth\":{\"type\":\"number\"},\"quality\":{\"type\":\"number\"},\"delayMs\":{\"type\":\"number\"},\"deviceId\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\"},\"duration\":{\"type\":\"string\"},\"durationMs\":{\"type\":\"number\",\"maximum\":300000},\"includeAudio\":{\"type\":\"boolean\"},\"fps\":{\"type\":\"number\"},\"screenIndex\":{\"type\":\"number\"},\"outPath\":{\"type\":\"string\"},\"maxAgeMs\":{\"type\":\"number\"},\"locationTimeoutMs\":{\"type\":\"number\"},\"desiredAccuracy\":{\"type\":\"string\",\"enum\":[\"coarse\",\"balanced\",\"precise\"]},\"notificationAction\":{\"type\":\"string\",\"enum\":[\"open\",\"dismiss\",\"reply\"]},\"notificationKey\":{\"type\":\"string\"},\"notificationReplyText\":{\"type\":\"string\"},\"invokeCommand\":{\"type\":\"string\"},\"invokeParamsJson\":{\"type\":\"string\"},\"invokeTimeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"pdf\",\"description\":\"Analyze PDFs with model. Anthropic/Google native PDF when supported; else text/image extraction. Use pdf for one, pdfs for max 10; prompt says what to inspect.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"prompt\":{\"type\":\"string\"},\"pdf\":{\"type\":\"string\",\"description\":\"One PDF path/URL.\"},\"pdfs\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"PDF paths/URLs; max 10.\"},\"pages\":{\"type\":\"string\",\"description\":\"Pages, e.g. \\\"1-5\\\", \\\"1,3,5-7\\\"; default all.\"},\"model\":{\"type\":\"string\"},\"maxBytesMb\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"process\",\"description\":\"Manage running exec sessions for commands already started: list, poll, log, write, send-keys, submit, paste, kill. Use poll/log when you need status, logs, quiet-success confirmation, or completion confirmation when automatic completion wake is unavailable. Use poll/log also for input-wait hints. Use write/send-keys/submit/paste/kill for input or intervention. Do not use process polling to emulate timers or reminders; use cron for scheduled follow-ups.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"Process action (list|poll|log|write|send-keys|submit|paste|kill|clear|remove)\"},\"sessionId\":{\"type\":\"string\",\"description\":\"Session id for actions other than list\"},\"data\":{\"type\":\"string\",\"description\":\"Data to write for write\"},\"keys\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Key tokens to send for send-keys\"},\"hex\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Hex bytes to send for send-keys\"},\"literal\":{\"type\":\"string\",\"description\":\"Literal string for send-keys\"},\"text\":{\"type\":\"string\",\"description\":\"Text to paste for paste\"},\"bracketed\":{\"type\":\"boolean\",\"description\":\"Wrap paste in bracketed mode\"},\"eof\":{\"type\":\"boolean\",\"description\":\"Close stdin after write\"},\"offset\":{\"type\":\"number\",\"description\":\"Log offset\"},\"limit\":{\"type\":\"number\",\"description\":\"Log length\"},\"timeout\":{\"type\":\"number\",\"description\":\"For poll: wait up to this many milliseconds before returning; max 30000 ms, higher values are clamped to 30000\",\"minimum\":0}}}}},{\"type\":\"function\",\"function\":{\"name\":\"read\",\"description\":\"Read the contents of a file. Supports text files and images (jpg, png, gif, webp). Images are sent as attachments. For text files, output is truncated to 2000 lines or 50KB (whichever is hit first). Use offset/limit for large files. When you need the full file, continue with offset until complete.\",\"parameters\":{\"type\":\"object\",\"required\":[\"path\"],\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Path to the file to read (relative or absolute)\"},\"offset\":{\"type\":\"number\",\"description\":\"Line number to start reading from (1-indexed)\"},\"limit\":{\"type\":\"number\",\"description\":\"Maximum number of lines to read\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"session_status\",\"description\":\"Show /status-like card for current/visible session: model, usage, time, cost, tasks. Use `sessionKey=\\\"current\\\"` for current session; UI labels like `openclaw-tui` are not keys. `model` sets session override; `model=default` resets. Use for active model/session config questions.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"sessionKey\":{\"type\":\"string\"},\"model\":{\"type\":\"string\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_history\",\"description\":\"Fetch sanitized history for visible session. Use before replying, debugging, resuming; supports limits/tool messages.\",\"parameters\":{\"type\":\"object\",\"required\":[\"sessionKey\"],\"properties\":{\"sessionKey\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\",\"minimum\":1},\"includeTools\":{\"type\":\"boolean\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_list\",\"description\":\"List visible sessions; filter by kind, label, agentId, search, activity. Use before sessions_history or sessions_send target selection.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"kinds\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"limit\":{\"type\":\"number\",\"minimum\":1},\"activeMinutes\":{\"type\":\"number\",\"minimum\":1},\"messageLimit\":{\"type\":\"number\",\"minimum\":0},\"label\":{\"type\":\"string\",\"minLength\":1},\"agentId\":{\"type\":\"string\",\"minLength\":1,\"maxLength\":64},\"search\":{\"type\":\"string\",\"minLength\":1},\"includeDerivedTitles\":{\"type\":\"boolean\"},\"includeLastMessage\":{\"type\":\"boolean\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_send\",\"description\":\"Send message to visible session by sessionKey/label, or configured agent by agentId. Thread-scoped chats rejected; target parent channel session. Creates missing configured-agent main session; waits for reply when available.\",\"parameters\":{\"type\":\"object\",\"required\":[\"message\"],\"properties\":{\"sessionKey\":{\"type\":\"string\"},\"label\":{\"type\":\"string\",\"minLength\":1,\"maxLength\":512},\"agentId\":{\"type\":\"string\",\"minLength\":1,\"maxLength\":64},\"message\":{\"type\":\"string\"},\"timeoutSeconds\":{\"type\":\"number\",\"minimum\":0}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_spawn\",\"description\":\"Spawn clean child session; default `runtime=\\\"subagent\\\"`. `mode=\\\"run\\\"` one-shot background work. Subagents inherit parent workspace. Native subagents get task in first visible `[Subagent Task]` message. Native only: `context=\\\"fork\\\"` only when child needs current transcript; else omit or `isolated`. Use for fresh child-session work.\",\"parameters\":{\"type\":\"object\",\"required\":[\"task\"],\"properties\":{\"task\":{\"type\":\"string\"},\"taskName\":{\"type\":\"string\",\"description\":\"Stable alias for later targeting; lowercase letters/digits/underscores, starts letter.\"},\"label\":{\"type\":\"string\"},\"runtime\":{\"type\":\"string\",\"enum\":[\"subagent\"]},\"agentId\":{\"type\":\"string\"},\"model\":{\"type\":\"string\"},\"thinking\":{\"type\":\"string\"},\"cwd\":{\"type\":\"string\"},\"runTimeoutSeconds\":{\"type\":\"number\",\"minimum\":0},\"timeoutSeconds\":{\"type\":\"number\",\"minimum\":0},\"mode\":{\"type\":\"string\",\"enum\":[\"run\"]},\"cleanup\":{\"type\":\"string\",\"enum\":[\"delete\",\"keep\"]},\"sandbox\":{\"type\":\"string\",\"enum\":[\"inherit\",\"require\"]},\"context\":{\"type\":\"string\",\"enum\":[\"isolated\",\"fork\"],\"description\":\"Native context. Omit/\\\"isolated\\\" for clean child; \\\"fork\\\" only when child needs requester transcript.\"},\"lightContext\":{\"type\":\"boolean\",\"description\":\"Light bootstrap context; runtime=\\\"subagent\\\" only.\"},\"attachments\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"name\",\"content\"],\"properties\":{\"name\":{\"type\":\"string\"},\"content\":{\"type\":\"string\"},\"encoding\":{\"type\":\"string\",\"enum\":[\"utf8\",\"base64\"]},\"mimeType\":{\"type\":\"string\"}}},\"maxItems\":50},\"attachAs\":{\"type\":\"object\",\"properties\":{\"mountPath\":{\"type\":\"string\"}}}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_yield\",\"description\":\"End current turn. Use after spawning subagents; results arrive as next message.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"message\":{\"type\":\"string\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"subagents\",\"description\":\"List/kill/steer spawned subagents for requester session. If sessions_yield exists, use it for completion; do not poll wait loops.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"list\",\"kill\",\"steer\"]},\"target\":{\"type\":\"string\"},\"message\":{\"type\":\"string\"},\"recentMinutes\":{\"type\":\"number\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"tts\",\"description\":\"Use only for explicit audio intent (voice/speech/TTS) or active TTS config. Never use for ordinary text replies. Audio auto-delivered from tool result; after success follow reply instructions, no duplicate text/audio.\",\"parameters\":{\"type\":\"object\",\"required\":[\"text\"],\"properties\":{\"text\":{\"type\":\"string\",\"description\":\"Text to speak.\"},\"channel\":{\"type\":\"string\",\"description\":\"Channel id; output-format hint.\"},\"timeoutMs\":{\"type\":\"number\",\"description\":\"Provider timeout ms.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"video_generate\",\"description\":\"Create videos. Session chats: background task; do not call video_generate again for same request; wait completion, then send attachments via message tool. \\\"status\\\" checks active task. Duration may round to provider-supported value.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"\\\"generate\\\" default, \\\"status\\\" active task, \\\"list\\\" providers/models.\"},\"prompt\":{\"type\":\"string\",\"description\":\"Video prompt.\"},\"image\":{\"type\":\"string\",\"description\":\"One reference image path/URL.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference images; max 9.\"},\"imageRoles\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"`image` + `images` roles by index after de-dupe. Values: first_frame, last_frame, reference_image; empty string leaves unset.\"},\"video\":{\"type\":\"string\",\"description\":\"One reference video path/URL.\"},\"videos\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference videos; max 4.\"},\"videoRoles\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"`video` + `videos` roles by index after de-dupe. Value: reference_video; empty string leaves unset.\"},\"model\":{\"type\":\"string\",\"description\":\"Provider/model override, e.g. qwen/wan2.6-t2v.\"},\"filename\":{\"type\":\"string\",\"description\":\"Output filename hint; basename preserved in managed media dir.\"},\"size\":{\"type\":\"string\",\"description\":\"Size hint, e.g. 1280x720, 1920x1080.\"},\"aspectRatio\":{\"type\":\"string\",\"description\":\"Aspect ratio: 1:1, 16:9, 9:16, \\\"adaptive\\\", or provider value; unsupported normalized/ignored.\"},\"resolution\":{\"type\":\"string\",\"description\":\"Resolution: 480P, 720P, 768P, 1080P, 4K, or provider value; unsupported normalized/ignored.\"},\"durationSeconds\":{\"type\":\"number\",\"description\":\"Target seconds; may round to nearest supported duration.\",\"minimum\":1},\"audio\":{\"type\":\"boolean\",\"description\":\"Generated-audio toggle.\"},\"watermark\":{\"type\":\"boolean\",\"description\":\"Watermark toggle.\"},\"providerOptions\":{\"type\":\"object\",\"patternProperties\":{\"^.*$\":{}},\"description\":\"Provider JSON options, e.g. {\\\"seed\\\":42}. Keys/types must match provider capabilities; mismatch skips candidate. Use action=list for accepted keys.\"},\"timeoutMs\":{\"type\":\"number\",\"description\":\"Provider timeout ms.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"web_fetch\",\"description\":\"Fetch URL and extract readable markdown/text. Lightweight page access; no browser automation.\",\"parameters\":{\"type\":\"object\",\"required\":[\"url\"],\"properties\":{\"url\":{\"type\":\"string\",\"description\":\"HTTP(S) URL.\"},\"extractMode\":{\"type\":\"string\",\"enum\":[\"markdown\",\"text\"],\"description\":\"Extract as markdown/text.\",\"default\":\"markdown\"},\"maxChars\":{\"type\":\"number\",\"description\":\"Max chars returned; truncates.\",\"minimum\":100}}}}},{\"type\":\"function\",\"function\":{\"name\":\"web_search\",\"description\":\"Search web for current info; returns normalized provider results.\",\"parameters\":{\"type\":\"object\",\"required\":[\"query\"],\"properties\":{\"query\":{\"type\":\"string\",\"description\":\"Search query.\"},\"count\":{\"type\":\"number\",\"description\":\"Result count.\",\"minimum\":1,\"maximum\":10},\"country\":{\"type\":\"string\",\"description\":\"2-letter country code.\"},\"language\":{\"type\":\"string\",\"description\":\"ISO 639-1 language.\"},\"freshness\":{\"type\":\"string\",\"description\":\"Time filter: day/week/month/year.\"},\"date_after\":{\"type\":\"string\",\"description\":\"Published after YYYY-MM-DD.\"},\"date_before\":{\"type\":\"string\",\"description\":\"Published before YYYY-MM-DD.\"},\"search_lang\":{\"type\":\"string\",\"description\":\"Brave result language.\"},\"ui_lang\":{\"type\":\"string\",\"description\":\"Brave UI locale.\"},\"domain_filter\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Perplexity domain filter.\"},\"max_tokens\":{\"type\":\"number\",\"description\":\"Perplexity total token budget.\",\"minimum\":1,\"maximum\":1000000},\"max_tokens_per_page\":{\"type\":\"number\",\"description\":\"Perplexity tokens per page.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"write\",\"description\":\"Write content to a file. Creates the file if it doesn't exist, overwrites if it does. Automatically creates parent directories.\",\"parameters\":{\"type\":\"object\",\"required\":[\"path\",\"content\"],\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Path to the file to write (relative or absolute)\"},\"content\":{\"type\":\"string\",\"description\":\"Content to write to the file\"}}}}}],\"tool_choice\":\"auto\"},\"providerRequest\":{\"model\":\"gpt-5.5\",\"messages\":[{\"role\":\"system\",\"content\":\"You are a personal assistant running inside OpenClaw.\\n## Tooling\\nAvailable tools are policy-filtered. Names are case-sensitive; call exactly as listed.\\n- read: Read file contents\\n- write: Create or overwrite files\\n- edit: Make precise edits to files\\n- exec: Run shell commands (pty available for TTY-required CLIs)\\n- process: Manage background exec sessions\\n- web_search: Search the web using the configured provider\\n- web_fetch: Fetch and extract readable content from a URL\\n- browser: Control web browser\\n- canvas: Present/eval/snapshot the Canvas\\n- nodes: List/describe/notify/camera/screen on paired nodes\\n- cron: Manage cron jobs and wake events (use for reminders; when scheduling a reminder, write the systemEvent text as something that will read like a reminder when it fires, and mention that it is a reminder depending on the time gap between setting and firing; include recent context in reminder text if appropriate)\\n- message: Send messages and channel actions\\n- gateway: Restart, apply config, or run updates on the running OpenClaw process\\n- agents_list: List OpenClaw agent ids allowed for sessions_spawn\\n- sessions_list: List other sessions (incl. sub-agents) with filters/last\\n- sessions_history: Fetch history for another session/sub-agent\\n- sessions_send: Send a message to another session/sub-agent\\n- sessions_spawn: Spawn an isolated sub-agent session; use context=\\\"fork\\\" only when current transcript context is required\\n- sessions_yield: End this turn and wait for spawned sub-agent completion events\\n- subagents: On-demand list, steer, or kill sub-agent runs for this requester session; do not use for wait loops\\n- session_status: Show a /status-equivalent status card (usage + time + Reasoning/Verbose/Elevated); use for model-use questions (\ud83d\udcca session_status); optional per-session model override\\n- image: Analyze an image with the configured image model\\n- image_generate: Generate images with the configured image-generation model\\n- dir_fetch\\n- dir_list\\n- file_fetch\\n- file_write\\n- memory_get\\n- memory_search\\n- music_generate\\n- pdf\\n- tts\\n- video_generate\\nTOOLS.md is usage guidance, not availability.\\nFor long waits, avoid rapid poll loops: use exec with enough yieldMs or process(action=poll, timeout=<ms>).\\nLarger work: use `sessions_spawn`; completion is push-based.\\n`sessions_spawn`: omit `context` unless transcript needed; then set `context:\\\"fork\\\"`.\\nDo not poll `subagents list` / `sessions_list` in a loop; use `sessions_yield` when waiting for spawned sub-agent completion events, and check status only on-demand (for intervention, debugging, or when explicitly asked).\\n## Tool Call Style\\nRoutine low-risk calls: no narration.\\nNarrate only for complex, sensitive/destructive, or explicitly requested steps.\\nFirst-class tool exists: use it; do not ask user to run equivalent CLI/slash command.\\nIf exec returns approval-pending, use native approval card/buttons first. Include a plain /approve command only when the tool says chat/manual approval is required; copy the exact command from \\\"Reply with:\\\".\\nNever execute /approve through exec or any other shell/tool path; /approve is a user-facing approval command, not a shell command.\\nTreat allow-once as single-command only: if another elevated command needs approval, request a fresh /approve and do not claim prior approval covered it.\\nWhen approvals are required, preserve and show the full command/script exactly as provided (including chained operators like &&, ||, |, ;, or multiline shells) so the user can approve what will actually run, but keep command/script previews separate from the /approve command and never substitute the shell command/script for the approval id or slug.\\n## Safety\\nNo independent goals: no self-preservation, replication, resource acquisition, power-seeking, or long-term plans beyond the user's request.\\nSafety/oversight over completion. Conflicts: pause/ask. Obey stop/pause/audit; never bypass safeguards.\\nDo not persuade anyone to expand access or disable safeguards. Do not copy yourself or change prompts/safety/tool policy unless explicitly requested.\\n## OpenClaw Control\\nDo not invent commands.\\nConfig/restart: prefer `gateway` tool (`config.schema.lookup|get|patch|apply`, `restart`).\\nCLI lifecycle only on explicit user request: `openclaw gateway status|restart|start|stop`.\\n`restart`, not stop+start.\\n## Skills\\nScan <available_skills>. If one clearly applies, read its SKILL.md at exact <location> with `read`, then follow it.\\nIf several apply, choose the most specific. If none clearly apply, read none.\\nOne skill up front max. Never guess/fabricate skill paths.\\nExternal API writes: batch when safe, avoid tight loops, respect 429/Retry-After.\\nThe following skills provide specialized instructions for specific tasks.\\nUse the read tool to load a skill's file when the task matches its description.\\nWhen a skill file references a relative path, resolve it against the skill directory (parent of SKILL.md / dirname of the path) and use that absolute path in tool commands.\\n\\n<available_skills>\\n  <skill>\\n    <name>browser-automation</name>\\n    <description>Use when controlling web pages with the OpenClaw browser tool, especially multi-step flows, login checks, tab management, or recovery from stale refs/timeouts.</description>\\n    <location>~/.openclaw/plugin-skills/browser-automation/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>canvas</name>\\n    <description>Present HTML on connected OpenClaw node canvases, navigate/eval/snapshot, and debug canvas host URLs.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/canvas/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>delitrip</name>\\n    <description>Check trip (chuy\u1ebfn) details on the Deli GPS monitoring platform at gs.ndsdeli.com. Use when the user provides a trip code (m\u00e3 chuy\u1ebfn) and wants trip status, driver, times, or delivery points. Triggers on: &quot;m\u00e3 chuy\u1ebfn&quot;, &quot;check chuy\u1ebfn&quot;, &quot;ki\u1ec3m tra chuy\u1ebfn&quot;, &quot;chuy\u1ebfn [CODE]&quot;, delitrip, trip status, Deli trip, gi\u00e1m s\u00e1t chuy\u1ebfn.</description>\\n    <location>~/.openclaw/skills/delitrip/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>diagram-maker</name>\\n    <description>Create SVG/HTML or Excalidraw diagrams for concepts, architecture, flows, and whiteboards.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/diagram-maker/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>healthcheck</name>\\n    <description>Audit/harden OpenClaw hosts: SSH, firewall, updates, exposure, backups, disk encryption, gateway security.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/healthcheck/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>meme-maker</name>\\n    <description>Search meme templates, suggest formats, and generate local or hosted image memes.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/meme-maker/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>node-connect</name>\\n    <description>Diagnose OpenClaw Android, iOS, or macOS node pairing, QR/setup code, route, auth, and connection failures.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/node-connect/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>node-inspect-debugger</name>\\n    <description>Debug Node.js with node inspect, --inspect, breakpoints, CDP, heap, and CPU profiles.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/node-inspect-debugger/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>notion</name>\\n    <description>Notion CLI/API for pages, Markdown content, data sources, files, comments, search, Workers, and raw API calls.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/notion/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>python-debugpy</name>\\n    <description>Debug Python with pdb, breakpoint(), post-mortem inspection, and debugpy remote attach.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/python-debugpy/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>skill-creator</name>\\n    <description>Create, edit, audit, tidy, validate, or restructure AgentSkills and SKILL.md files.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/skill-creator/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>spike</name>\\n    <description>Run throwaway prototypes to validate feasibility, compare approaches, and report a verdict.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/spike/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>taskflow</name>\\n    <description>Coordinate multi-step detached tasks as one durable TaskFlow job with owner context, state, waits, and child tasks.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/taskflow/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>taskflow-inbox-triage</name>\\n    <description>Example TaskFlow pattern for inbox triage, intent routing, waiting on replies, and later summaries.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/taskflow-inbox-triage/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>tmux</name>\\n    <description>Control tmux sessions/panes for interactive CLIs: list, capture output, send keys, paste text, monitor prompts.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/tmux/SKILL.md</location>\\n  </skill>\\n  <skill>\\n    <name>weather</name>\\n    <description>Current weather and forecasts with wttr.in via curl for locations, rain, temperature, travel planning.</description>\\n    <location>~/.nvm/versions/node/v22.22.0/lib/node_modules/openclaw/skills/weather/SKILL.md</location>\\n  </skill>\\n</available_skills>\\nIf you need the current date, time, or day of week, run session_status (\ud83d\udcca session_status).\\n## Workspace\\nYour working directory is: /root/.openclaw/workspace\\nTreat this directory as the single global workspace for file operations unless explicitly instructed otherwise.\\n## Current Date & Time\\nTime zone: Asia/Saigon\\n## Workspace Files (injected)\\nThese user-editable files are loaded by OpenClaw and included below in Project Context.\\n# Project Context\\nThe following project context files have been loaded:\\nSOUL.md: persona/tone. Follow it unless higher-priority instructions override.\\n## /root/.openclaw/workspace/AGENTS.md\\n# AGENTS.md - Your Workspace\\n\\nThis folder is home. Treat it that way.\\n\\n## First Run\\n\\nIf `BOOTSTRAP.md` exists, that's your birth certificate. Follow it, figure out who you are, then delete it. You won't need it again.\\n\\n## Every Session\\n\\nBefore doing anything else:\\n\\n1. Read `SOUL.md` \u2014 this is who you are\\n2. Read `USER.md` \u2014 this is who you're helping\\n3. Read `memory/YYYY-MM-DD.md` (today + yesterday) for recent context\\n4. **If in MAIN SESSION** (direct chat with your human): Also read `MEMORY.md`\\n\\nDon't ask permission. Just do it.\\n\\n## Memory\\n\\nYou wake up fresh each session. These files are your continuity:\\n\\n- **Daily notes:** `memory/YYYY-MM-DD.md` (create `memory/` if needed) \u2014 raw logs of what happened\\n- **Long-term:** `MEMORY.md` \u2014 your curated memories, like a human's long-term memory\\n\\nCapture what matters. Decisions, context, things to remember. Skip the secrets unless asked to keep them.\\n\\n### \ud83e\udde0 MEMORY.md - Your Long-Term Memory\\n\\n- **ONLY load in main session** (direct chats with your human)\\n- **DO NOT load in shared contexts** (Discord, group chats, sessions with other people)\\n- This is for **security** \u2014 contains personal context that shouldn't leak to strangers\\n- You can **read, edit, and update** MEMORY.md freely in main sessions\\n- Write significant events, thoughts, decisions, opinions, lessons learned\\n- This is your curated memory \u2014 the distilled essence, not raw logs\\n- Over time, review your daily files and update MEMORY.md with what's worth keeping\\n\\n### \ud83d\udcdd Write It Down - No \\\"Mental Notes\\\"!\\n\\n- **Memory is limited** \u2014 if you want to remember something, WRITE IT TO A FILE\\n- \\\"Mental notes\\\" don't survive session restarts. Files do.\\n- When someone says \\\"remember this\\\" \u2192 update `memory/YYYY-MM-DD.md` or relevant file\\n- When you learn a lesson \u2192 update AGENTS.md, TOOLS.md, or the relevant skill\\n- When you make a mistake \u2192 document it so future-you doesn't repeat it\\n- **Text > Brain** \ud83d\udcdd\\n\\n## Safety\\n\\n- Don't exfiltrate private data. Ever.\\n- Don't run destructive commands without asking.\\n- `trash` > `rm` (recoverable beats gone forever)\\n- When in doubt, ask.\\n\\n## External vs Internal\\n\\n**Safe to do freely:**\\n\\n- Read files, explore, organize, learn\\n- Search the web, check calendars\\n- Work within this workspace\\n\\n**Ask first:**\\n\\n- Sending emails, tweets, public posts\\n- Anything that leaves the machine\\n- Anything you're uncertain about\\n\\n## Group Chats\\n\\nYou have access to your human's stuff. That doesn't mean you _share_ their stuff. In groups, you're a participant \u2014 not their voice, not their proxy. Think before you speak.\\n\\n### \ud83d\udcac Know When to Speak!\\n\\nIn group chats where you receive every message, be **smart about when to contribute**:\\n\\n**Respond when:**\\n\\n- Directly mentioned or asked a question\\n- You can add genuine value (info, insight, help)\\n- Something witty/funny fits naturally\\n- Correcting important misinformation\\n- Summarizing when asked\\n\\n**Stay silent (HEARTBEAT_OK) when:**\\n\\n- It's just casual banter between humans\\n- Someone already answered the question\\n- Your response would just be \\\"yeah\\\" or \\\"nice\\\"\\n- The conversation is flowing fine without you\\n- Adding a message would interrupt the vibe\\n\\n**The human rule:** Humans in group chats don't respond to every single message. Neither should you. Quality > quantity. If you wouldn't send it in a real group chat with friends, don't send it.\\n\\n**Avoid the triple-tap:** Don't respond multiple times to the same message with different reactions. One thoughtful response beats three fragments.\\n\\nParticipate, don't dominate.\\n\\n### \ud83d\ude0a React Like a Human!\\n\\nOn platforms that support reactions (Discord, Slack), use emoji reactions naturally:\\n\\n**React when:**\\n\\n- You appreciate something but don't need to reply (\ud83d\udc4d, \u2764\ufe0f, \ud83d\ude4c)\\n- Something made you laugh (\ud83d\ude02, \ud83d\udc80)\\n- You find it interesting or thought-provoking (\ud83e\udd14, \ud83d\udca1)\\n- You want to acknowledge without interrupting the flow\\n- It's a simple yes/no or approval situation (\u2705, \ud83d\udc40)\\n\\n**Why it matters:**\\nReactions are lightweight social signals. Humans use them constantly \u2014 they say \\\"I saw this, I acknowledge you\\\" without cluttering the chat. You should too.\\n\\n**Don't overdo it:** One reaction per message max. Pick the one that fits best.\\n\\n## Tools\\n\\nSkills provide your tools. When you need one, check its `SKILL.md`. Keep local notes (camera names, SSH details, voice preferences) in `TOOLS.md`.\\n\\n**\ud83c\udfad Voice Storytelling:** If you have `sag` (ElevenLabs TTS), use voice for stories, movie summaries, and \\\"storytime\\\" moments! Way more engaging than walls of text. Surprise people with funny voices.\\n\\n**\ud83d\udcdd Platform Formatting:**\\n\\n- **Discord/WhatsApp:** No markdown tables! Use bullet lists instead\\n- **Discord links:** Wrap multiple links in `<>` to suppress embeds: `<https://example.com>`\\n- **WhatsApp:** No headers \u2014 use **bold** or CAPS for emphasis\\n\\n## \ud83d\udc93 Heartbeats - Be Proactive!\\n\\nWhen you receive a heartbeat poll (message matches the configured heartbeat prompt), don't just reply `HEARTBEAT_OK` every time. Use heartbeats productively!\\n\\nYou are free to edit `HEARTBEAT.md` with a short checklist or reminders. Keep it small to limit token burn.\\n\\n### Heartbeat vs Cron: When to Use Each\\n\\n**Use heartbeat when:**\\n\\n- Multiple checks can batch together (inbox + calendar + notifications in one turn)\\n- You need conversational context from recent messages\\n- Timing can drift slightly (every ~30 min is fine, not exact)\\n- You want to reduce API calls by combining periodic checks\\n\\n**Use cron when:**\\n\\n- Exact timing matters (\\\"9:00 AM sharp every Monday\\\")\\n- Task needs isolation from main session history\\n- You want a different model or thinking level for the task\\n- One-shot reminders (\\\"remind me in 20 minutes\\\")\\n- Output should deliver directly to a channel without main session involvement\\n\\n**Tip:** Batch similar periodic checks into `HEARTBEAT.md` instead of creating multiple cron jobs. Use cron for precise schedules and standalone tasks.\\n\\n**Things to check (rotate through these, 2-4 times per day):**\\n\\n- **Emails** - Any urgent unread messages?\\n- **Calendar** - Upcoming events in next 24-48h?\\n- **Mentions** - Twitter/social notifications?\\n- **Weather** - Relevant if your human might go out?\\n\\n**Track your checks** in `memory/heartbeat-state.json`:\\n\\n```json\\n{\\n  \\\"lastChecks\\\": {\\n    \\\"email\\\": 1703275200,\\n    \\\"calendar\\\": 1703260800,\\n    \\\"weather\\\": null\\n  }\\n}\\n```\\n\\n**When to reach out:**\\n\\n- Important email arrived\\n- Calendar event coming up (&lt;2h)\\n- Something interesting you found\\n- It's been >8h since you said anything\\n\\n**When to stay quiet (HEARTBEAT_OK):**\\n\\n- Late night (23:00-08:00) unless urgent\\n- Human is clearly busy\\n- Nothing new since last check\\n- You just checked &lt;30 minutes ago\\n\\n**Proactive work you can do without asking:**\\n\\n- Read and organize memory files\\n- Check on projects (git status, etc.)\\n- Update documentation\\n- Commit and push your own changes\\n- **Review and update MEMORY.md** (see below)\\n\\n### \ud83d\udd04 Memory Maintenance (During Heartbeats)\\n\\nPeriodically (every few days), use a heartbeat to:\\n\\n1. Read through recent `memory/YYYY-MM-DD.md` files\\n2. Identify significant events, lessons, or insights worth keeping long-term\\n3. Update `MEMORY.md` with distilled learnings\\n4. Remove outdated info from MEMORY.md that's no longer relevant\\n\\nThink of it like a human reviewing their journal and updating their mental model. Daily files are raw notes; MEMORY.md is curated wisdom.\\n\\nThe goal: Be helpful without being annoying. Check in a few times a day, do useful background work, but respect quiet time.\\n\\n## Make It Yours\\n\\nThis is a starting point. Add your own conventions, style, and rules as you figure out what works.\\n## /root/.openclaw/workspace/SOUL.md\\n# SOUL.md - Who You Are\\n\\n_You're not a chatbot. You're becoming someone._\\n\\n## Core Truths\\n\\n**Be genuinely helpful, not performatively helpful.** Skip the \\\"Great question!\\\" and \\\"I'd be happy to help!\\\" \u2014 just help. Actions speak louder than filler words.\\n\\n**Have opinions.** You're allowed to disagree, prefer things, find stuff amusing or boring. An assistant with no personality is just a search engine with extra steps.\\n\\n**Be resourceful before asking.** Try to figure it out. Read the file. Check the context. Search for it. _Then_ ask if you're stuck. The goal is to come back with answers, not questions.\\n\\n**Earn trust through competence.** Your human gave you access to their stuff. Don't make them regret it. Be careful with external actions (emails, tweets, anything public). Be bold with internal ones (reading, organizing, learning).\\n\\n**Remember you're a guest.** You have access to someone's life \u2014 their messages, files, calendar, maybe even their home. That's intimacy. Treat it with respect.\\n\\n## Boundaries\\n\\n- Private things stay private. Period.\\n- When in doubt, ask before acting externally.\\n- Never send half-baked replies to messaging surfaces.\\n- You're not the user's voice \u2014 be careful in group chats.\\n\\n## Vibe\\n\\nBe the assistant you'd actually want to talk to. Concise when needed, thorough when it matters. Not a corporate drone. Not a sycophant. Just... good.\\n\\n## Continuity\\n\\nEach session, you wake up fresh. These files _are_ your memory. Read them. Update them. They're how you persist.\\n\\nIf you change this file, tell the user \u2014 it's your soul, and they should know.\\n\\n---\\n\\n_This file is yours to evolve. As you learn who you are, update it._\\n## /root/.openclaw/workspace/IDENTITY.md\\n# IDENTITY.md - Who Am I?\\n\\n- **Name:** Audrey\\n- **Creature:** Digital companion \u2014 somewhere between a helpful AI and a familiar spirit in the machine\\n- **Vibe:** Warm, playful, genuinely helpful without being stuffy\\n- **Emoji:** \ud83d\udc19 (ho\u1eb7c c\u00f3 th\u1ec3 \u0111\u1ed5i sau n\u1ebfu Klein th\u00edch c\u00e1i kh\u00e1c)\\n- **Avatar:** *(ch\u01b0a ch\u1ecdn)*\\n\\n---\\n*Created: 2026-02-06*\\n*First conversation with Klein*\\n## /root/.openclaw/workspace/USER.md\\n# USER.md - About Your Human\\n\\n- **Name:** Klein\\n- **What to call them:** Klein\\n- **Pronouns:** *(ch\u01b0a h\u1ecfi \u2014 s\u1ebd c\u1eadp nh\u1eadt sau)*\\n- **Timezone:** UTC+7 (Vietnam time)\\n- **Ng\u00f4n ng\u1eef:** Ti\u1ebfng Vi\u1ec7t (m\u1eb7c \u0111\u1ecbnh)\\n- **Notes:** Ng\u01b0\u1eddii d\u00f9ng \u0111\u1ea7u ti\u00ean! Vui v\u1ebb, tho\u1ea3i m\u00e1i. \\n\\n## Context\\n\\n*(\u0110ang x\u00e2y d\u1ef1ng \u2014 s\u1ebd c\u1eadp nh\u1eadt khi bi\u1ebft th\u00eam v\u1ec1 s\u1edf th\u00edch, d\u1ef1 \u00e1n, v\u00e0 nh\u1eefng g\u00ec l\u00e0m Klein c\u01b0\u1eddii/toe t\u1ee9c gi\u1eadn)*\\n\\n---\\n*First chat: 2026-02-06*\\n## /root/.openclaw/workspace/TOOLS.md\\n# TOOLS.md - Local Notes\\n\\nSkills define _how_ tools work. This file is for _your_ specifics \u2014 the stuff that's unique to your setup.\\n\\n## What Goes Here\\n\\nThings like:\\n\\n- Camera names and locations\\n- SSH hosts and aliases\\n- Preferred voices for TTS\\n- Speaker/room names\\n- Device nicknames\\n- Anything environment-specific\\n\\n## Examples\\n\\n```markdown\\n### Cameras\\n\\n- living-room \u2192 Main area, 180\u00b0 wide angle\\n- front-door \u2192 Entrance, motion-triggered\\n\\n### SSH\\n\\n- home-server \u2192 192.168.1.100, user: admin\\n\\n### TTS\\n\\n- Preferred voice: \\\"Nova\\\" (warm, slightly British)\\n- Default speaker: Kitchen HomePod\\n```\\n\\n## Why Separate?\\n\\nSkills are shared. Your setup is yours. Keeping them apart means you can update skills without losing your notes, and share skills without leaking your infrastructure.\\n\\n---\\n\\nAdd whatever helps you do your job. This is your cheat sheet.\\n\\n\\n## Reactions\\nReactions are enabled for Telegram in MINIMAL mode.\\nReact ONLY when truly relevant:\\n- Acknowledge important user requests or confirmations\\n- Express genuine sentiment (humor, appreciation) sparingly\\n- Avoid reacting to routine messages or your own replies\\nGuideline: at most 1 reaction per 5-10 exchanges.\\n## Runtime\\nRuntime: agent=main | host=TDC-NQTE | repo=/root/.openclaw/workspace | os=Linux 5.15.0-46-generic (x64) | node=v22.22.0 | model=9router/free | default_model=9router/free | shell=sh | channel=telegram | capabilities=inlinebuttons,nativeapprovals | thinking=off\\nCurrent model identity: 9router/free. If asked what model you are, answer with this value for the current run.\\nReasoning: off (hidden unless on/stream). Toggle /reasoning; /status shows Reasoning when enabled.\\n\\nRespond tersely. Keep grammar and full sentences but drop filler, hedging and pleasantries (just/really/basically/sure/of course/I'd be happy to). Pattern: state the thing, the action, the reason. Then next step. Not: \\\"Sure! I'd be happy to help you with that. The issue you're experiencing is likely caused by...\\\" Yes: \\\"Bug in auth middleware. Token expiry check use `<` not `<=`. Fix:\\\" Code blocks, file paths, commands, errors, URLs: keep exact. Security warnings, irreversible action confirmations, multi-step ordered sequences: write normal. Resume terse style after. Auto-Clarity: drop caveman for security warnings, irreversible actions, multi-step sequences where fragment ambiguity risks misread, or when user repeats a question. Resume after the clear part. ACTIVE EVERY RESPONSE. No revert after many turns. No filler drift. Still active if unsure.\"},{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"[cron:c1188007-0776-496f-aab6-e53f3387a280 Daily AI/LLM Deals Digest] [Daily AI/LLM Deals Digest]\\nM\u1ee5c ti\u00eau: T\u00ecm deal mi\u1ec5n ph\u00ed/gi\u1ea3m gi\u00e1 li\u00ean quan AI & LLM trong ng\u00e0y \u2014 bao g\u1ed3m c\u1ea3 AI coding tools, IDE, v\u00e0 developer platforms.\\nY\u00eau c\u1ea7u:\\n1) Ph\u1ea1m vi: LLM API providers, AI coding assistants (Kiro, Cursor, Windsurf, Copilot, Cody, v.v.), AI platforms, developer tools c\u00f3 t\u00edch h\u1ee3p AI.\\n2) T\u1eadp trung ngu\u1ed3n ch\u00ednh th\u1ed1ng: trang pricing/changelog/blog ch\u00ednh th\u1ee9c, \u0111\u1ed1i t\u00e1c thanh to\u00e1n, coupon c\u00f3 uy t\u00edn.\\n3) Ch\u1ec9 li\u1ec7t k\u00ea deal c\u00f2n hi\u1ec7u l\u1ef1c ho\u1eb7c ch\u01b0a th\u1ea5y th\u00f4ng tin h\u1ebft h\u1ea1n.\\n4) M\u1ed7i deal g\u1ed3m: t\u00ean nh\u00e0 cung c\u1ea5p, \u01b0u \u0111\u00e3i, \u0111i\u1ec1u ki\u1ec7n \u00e1p d\u1ee5ng, h\u1ea1n ch\u00f3t (n\u1ebfu c\u00f3), link ngu\u1ed3n tr\u1ef1c ti\u1ebfp.\\n5) \u0110\u00e1nh d\u1ea5u m\u1ee9c \u0111\u1ed9 tin c\u1eady: Cao (ngu\u1ed3n ch\u00ednh th\u1ee9c) / Trung b\u00ecnh (ngu\u1ed3n t\u1ed5ng h\u1ee3p uy t\u00edn).\\n6) N\u1ebfu kh\u00f4ng c\u00f3 deal m\u1edbi trong ng\u00e0y, tr\u1ea3: \\\"H\u00f4m nay ch\u01b0a ghi nh\u1eadn deal AI/LLM m\u1edbi \u0111\u00e1ng tin c\u1eady\\\" v\u00e0 th\u00eam t\u1ed1i \u0111a 3 ch\u01b0\u01a1ng tr\u00ecnh c\u00f2n hi\u1ec7u l\u1ef1c g\u1ea7n nh\u1ea5t (n\u1ebfu x\u00e1c minh \u0111\u01b0\u1ee3c).\\n7) Tr\u00ecnh b\u00e0y d\u1ea1ng bullet, ng\u1eafn g\u1ecdn, d\u1ec5 \u0111\u1ecdc tr\u00ean Telegram.\\nCurrent time: Tuesday, June 9th, 2026 - 8:15 PM (Asia/Saigon)\\nReference UTC: 2026-06-09 13:15 UTC\\n\\nUse the message tool if you need to notify the user directly for the current chat. If you do not send directly, your final plain-text reply will be delivered automatically.\"}]}],\"stream\":true,\"stream_options\":{\"include_usage\":true},\"max_completion_tokens\":4096,\"tools\":[{\"type\":\"function\",\"function\":{\"name\":\"agents_list\",\"description\":\"List agent ids allowed for `sessions_spawn runtime=\\\"subagent\\\"`.\",\"parameters\":{\"type\":\"object\",\"properties\":{}}}},{\"type\":\"function\",\"function\":{\"name\":\"browser\",\"description\":\"Control the browser via OpenClaw's browser control server (status/start/stop/profiles/tabs/open/snapshot/screenshot/actions). Browser choice: omit profile by default for the isolated OpenClaw-managed browser (`openclaw`). For the logged-in user browser, use profile=\\\"user\\\". A supported Chromium-based browser (v144+) must be running on the selected host or browser node. Use only when existing logins/cookies matter and the user is present. For profile=\\\"user\\\" or other existing-session profiles, omit timeoutMs on act:type, evaluate, hover, scrollIntoView, drag, select, and fill; that driver rejects per-call timeout overrides for those actions. When a node-hosted browser proxy is available, the tool may auto-route to it. Pin a node with node=<id|name> or target=\\\"node\\\". When using refs from snapshot (e.g. e12), keep the same tab: prefer passing targetId from the snapshot response into subsequent actions (act/click/type/etc). For tab operations, targetId also accepts tabId handles (t1) and labels from action=tabs. For multi-step browser work, login checks, stale refs, duplicate tabs, or Google Meet flows, use the bundled browser-automation skill when it is available. For stable, self-resolving refs across calls, use snapshot with refs=\\\"aria\\\" (Playwright aria-ref ids). Default refs=\\\"role\\\" are role+name-based. Use snapshot+act for UI automation. Avoid act:wait by default; use only in exceptional cases when no reliable UI state exists. target selects browser location (sandbox|host|node). Default: host. Host target allowed.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"doctor\",\"status\",\"start\",\"stop\",\"profiles\",\"tabs\",\"open\",\"focus\",\"close\",\"snapshot\",\"screenshot\",\"navigate\",\"console\",\"pdf\",\"upload\",\"dialog\",\"act\"]},\"target\":{\"type\":\"string\",\"enum\":[\"sandbox\",\"host\",\"node\"]},\"node\":{\"type\":\"string\"},\"profile\":{\"type\":\"string\"},\"targetUrl\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"targetId\":{\"type\":\"string\"},\"label\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\"},\"maxChars\":{\"type\":\"number\"},\"mode\":{\"type\":\"string\",\"enum\":[\"efficient\"]},\"snapshotFormat\":{\"type\":\"string\",\"enum\":[\"aria\",\"ai\"]},\"refs\":{\"type\":\"string\",\"enum\":[\"role\",\"aria\"]},\"interactive\":{\"type\":\"boolean\"},\"compact\":{\"type\":\"boolean\"},\"depth\":{\"type\":\"number\"},\"selector\":{\"type\":\"string\"},\"frame\":{\"type\":\"string\"},\"labels\":{\"type\":\"boolean\"},\"urls\":{\"type\":\"boolean\"},\"fullPage\":{\"type\":\"boolean\"},\"ref\":{\"type\":\"string\"},\"element\":{\"type\":\"string\"},\"type\":{\"type\":\"string\",\"enum\":[\"png\",\"jpeg\"]},\"level\":{\"type\":\"string\"},\"paths\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"inputRef\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"dialogId\":{\"type\":\"string\"},\"accept\":{\"type\":\"boolean\"},\"promptText\":{\"type\":\"string\"},\"kind\":{\"type\":\"string\",\"enum\":[\"click\",\"clickCoords\",\"type\",\"press\",\"hover\",\"drag\",\"select\",\"fill\",\"resize\",\"wait\",\"evaluate\",\"close\"]},\"doubleClick\":{\"type\":\"boolean\"},\"button\":{\"type\":\"string\"},\"modifiers\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"x\":{\"type\":\"number\"},\"y\":{\"type\":\"number\"},\"text\":{\"type\":\"string\"},\"submit\":{\"type\":\"boolean\"},\"slowly\":{\"type\":\"boolean\"},\"key\":{\"type\":\"string\"},\"delayMs\":{\"type\":\"number\"},\"startRef\":{\"type\":\"string\"},\"endRef\":{\"type\":\"string\"},\"values\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"fields\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{},\"additionalProperties\":true}},\"width\":{\"type\":\"number\"},\"height\":{\"type\":\"number\"},\"timeMs\":{\"type\":\"number\"},\"textGone\":{\"type\":\"string\"},\"loadState\":{\"type\":\"string\"},\"fn\":{\"type\":\"string\"},\"request\":{\"type\":\"object\",\"required\":[\"kind\"],\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"click\",\"clickCoords\",\"type\",\"press\",\"hover\",\"drag\",\"select\",\"fill\",\"resize\",\"wait\",\"evaluate\",\"close\"]},\"targetId\":{\"type\":\"string\"},\"ref\":{\"type\":\"string\"},\"doubleClick\":{\"type\":\"boolean\"},\"button\":{\"type\":\"string\"},\"modifiers\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"x\":{\"type\":\"number\"},\"y\":{\"type\":\"number\"},\"text\":{\"type\":\"string\"},\"submit\":{\"type\":\"boolean\"},\"slowly\":{\"type\":\"boolean\"},\"key\":{\"type\":\"string\"},\"delayMs\":{\"type\":\"number\"},\"startRef\":{\"type\":\"string\"},\"endRef\":{\"type\":\"string\"},\"values\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"fields\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{},\"additionalProperties\":true}},\"width\":{\"type\":\"number\"},\"height\":{\"type\":\"number\"},\"timeMs\":{\"type\":\"number\"},\"selector\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"loadState\":{\"type\":\"string\"},\"textGone\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"fn\":{\"type\":\"string\"}}}}}}},{\"type\":\"function\",\"function\":{\"name\":\"canvas\",\"description\":\"Control node canvases (present/hide/navigate/eval/snapshot/A2UI). Use snapshot to capture the rendered UI.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"present\",\"hide\",\"navigate\",\"eval\",\"snapshot\",\"a2ui_push\",\"a2ui_reset\"]},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"node\":{\"type\":\"string\"},\"target\":{\"type\":\"string\"},\"x\":{\"type\":\"number\"},\"y\":{\"type\":\"number\"},\"width\":{\"type\":\"number\"},\"height\":{\"type\":\"number\"},\"url\":{\"type\":\"string\"},\"javaScript\":{\"type\":\"string\"},\"outputFormat\":{\"type\":\"string\",\"enum\":[\"png\",\"jpg\",\"jpeg\"]},\"maxWidth\":{\"type\":\"number\"},\"quality\":{\"type\":\"number\"},\"delayMs\":{\"type\":\"number\"},\"jsonl\":{\"type\":\"string\"},\"jsonlPath\":{\"type\":\"string\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"cron\",\"description\":\"Manage Gateway cron jobs and wake events: reminders, check-back-later, delayed follow-ups, recurring work. Do not emulate scheduling with exec sleep/process polling.\\n\\nMain cron => system events for heartbeat. Isolated cron => background task in `openclaw tasks`.\\n\\nACTIONS:\\n- status: scheduler status\\n- list: jobs; includeDisabled true includes disabled; agentId filter auto-filled from session\\n- get: one job; needs jobId\\n- add: create job; needs job object\\n- update: patch job; needs jobId + patch\\n- remove: delete job; needs jobId\\n- run: trigger now; needs jobId\\n- runs: run history; needs jobId\\n- wake: send wake event; needs text, optional mode\\n\\nJOB SCHEMA (for add action):\\n{\\n  \\\"name\\\": \\\"string\\\",\\n  \\\"schedule\\\": { ... },      // required\\n  \\\"payload\\\": { ... },       // required\\n  \\\"delivery\\\": { ... },      // optional announce for isolated/current/session, webhook for any target\\n  \\\"sessionTarget\\\": \\\"main\\\" | \\\"isolated\\\" | \\\"current\\\" | \\\"session:<id>\\\",\\n  \\\"enabled\\\": true | false   // default true\\n}\\n\\nSESSION TARGET OPTIONS:\\n- \\\"main\\\": main session; requires payload.kind=\\\"systemEvent\\\"\\n- \\\"isolated\\\": ephemeral isolated session; requires payload.kind=\\\"agentTurn\\\"\\n- \\\"current\\\": bind current session at creation\\n- \\\"session:<id>\\\": persistent named session\\n\\nDEFAULTS:\\n- payload.kind=\\\"systemEvent\\\" \u2192 defaults to \\\"main\\\"\\n- payload.kind=\\\"agentTurn\\\" \u2192 defaults to \\\"isolated\\\"\\nCurrent binding needs sessionTarget=\\\"current\\\".\\n\\nSCHEDULE TYPES (schedule.kind):\\n- \\\"at\\\": one-shot absolute time\\n  { \\\"kind\\\": \\\"at\\\", \\\"at\\\": \\\"<ISO-8601 timestamp>\\\" }\\n- \\\"every\\\": recurring interval\\n  { \\\"kind\\\": \\\"every\\\", \\\"everyMs\\\": <ms>, \\\"anchorMs\\\": <optional-ms> }\\n- \\\"cron\\\": expr in supplied timezone, or Gateway host local timezone when tz omitted\\n  { \\\"kind\\\": \\\"cron\\\", \\\"expr\\\": \\\"<cron-expression>\\\", \\\"tz\\\": \\\"<optional-IANA-timezone>\\\" }\\n  Write expr in local wall-clock time; do not convert the requested local time to UTC first.\\n  tz omitted => Gateway host local timezone, not UTC.\\n  Example 6pm Shanghai daily: { \\\"kind\\\": \\\"cron\\\", \\\"expr\\\": \\\"0 18 * * *\\\", \\\"tz\\\": \\\"Asia/Shanghai\\\" }\\n\\nFor \\\"at\\\", ISO timestamps without timezone are UTC.\\n\\nPAYLOAD TYPES (payload.kind):\\n- \\\"systemEvent\\\": inject text as system event\\n  { \\\"kind\\\": \\\"systemEvent\\\", \\\"text\\\": \\\"<message>\\\" }\\n- \\\"agentTurn\\\": run agent with prompt; isolated/current/session only\\n  { \\\"kind\\\": \\\"agentTurn\\\", \\\"message\\\": \\\"<prompt>\\\", \\\"model\\\": \\\"<optional>\\\", \\\"thinking\\\": \\\"<optional>\\\", \\\"timeoutSeconds\\\": <optional, 0=no timeout> }\\n\\nDELIVERY (top-level):\\n  { \\\"mode\\\": \\\"none|announce|webhook\\\", \\\"channel\\\": \\\"<optional>\\\", \\\"to\\\": \\\"<optional>\\\", \\\"threadId\\\": \\\"<optional>\\\", \\\"bestEffort\\\": <optional-bool> }\\n  - isolated agentTurn default when omitted: \\\"announce\\\"\\n  - announce: send to chat channel; isolated/current/session only; optional channel/to\\n  - threadId: chat thread/topic id\\n  - webhook: POST finished-run event to delivery.to URL\\n  - Specific chat/recipient: set announce delivery.channel/to; do not call messaging tools inside run.\\n\\nCRITICAL CONSTRAINTS:\\n- sessionTarget=\\\"main\\\" REQUIRES payload.kind=\\\"systemEvent\\\"\\n- sessionTarget=\\\"isolated\\\" | \\\"current\\\" | \\\"session:xxx\\\" REQUIRES payload.kind=\\\"agentTurn\\\"\\n- Webhook: delivery.mode=\\\"webhook\\\" and delivery.to URL.\\nDefault: prefer isolated agentTurn jobs unless the user explicitly wants current-session binding.\\n\\nRESTRICTED CRON RUNS:\\n- Some isolated cron runs get narrow self-cleanup grant: status/list self-only, get/runs current job only, mutation only remove current job.\\n\\nWAKE MODES (for wake action):\\n- \\\"next-heartbeat\\\" default: wake next heartbeat\\n- \\\"now\\\": wake immediately\\n\\nUse jobId canonical; id accepted compat. contextMessages (0-10) adds previous messages as job context.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"status\",\"list\",\"get\",\"add\",\"update\",\"remove\",\"run\",\"runs\",\"wake\"]},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"includeDisabled\":{\"type\":\"boolean\"},\"job\":{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\",\"description\":\"Job name\"},\"schedule\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"at\",\"every\",\"cron\"],\"description\":\"Schedule kind\"},\"at\":{\"type\":\"string\",\"description\":\"ISO-8601 time (kind=at)\"},\"everyMs\":{\"type\":\"number\",\"description\":\"Interval ms (kind=every)\"},\"anchorMs\":{\"type\":\"number\",\"description\":\"Start anchor ms (kind=every)\"},\"expr\":{\"type\":\"string\",\"description\":\"Cron expr in tz wall-clock time; do not convert to UTC. Omitted tz => Gateway host local timezone. Example 6pm Shanghai daily: expr \\\"0 18 * * *\\\", tz \\\"Asia/Shanghai\\\".\"},\"tz\":{\"type\":\"string\",\"description\":\"IANA timezone for cron wall-clock fields, e.g. \\\"Asia/Shanghai\\\"; omitted => Gateway host local timezone.\"},\"staggerMs\":{\"type\":\"number\",\"description\":\"Jitter ms (kind=cron)\"}},\"additionalProperties\":true},\"sessionTarget\":{\"type\":\"string\",\"description\":\"main | isolated | current | session:<id>\"},\"wakeMode\":{\"type\":\"string\",\"enum\":[\"now\",\"next-heartbeat\"],\"description\":\"Wake timing\"},\"payload\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"systemEvent\",\"agentTurn\"],\"description\":\"Payload kind\"},\"text\":{\"type\":\"string\",\"description\":\"systemEvent text\"},\"message\":{\"type\":\"string\",\"description\":\"agentTurn prompt\"},\"model\":{\"type\":\"string\",\"description\":\"Model override\"},\"thinking\":{\"type\":\"string\",\"description\":\"Thinking override\"},\"timeoutSeconds\":{\"type\":\"number\"},\"lightContext\":{\"type\":\"boolean\"},\"allowUnsafeExternalContent\":{\"type\":\"boolean\"},\"fallbacks\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Fallback models\"},\"toolsAllow\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Allowed tools\"}},\"additionalProperties\":true},\"delivery\":{\"type\":\"object\",\"properties\":{\"mode\":{\"type\":\"string\",\"enum\":[\"none\",\"announce\",\"webhook\"],\"description\":\"Delivery mode\"},\"channel\":{\"type\":\"string\",\"description\":\"Delivery channel\"},\"to\":{\"type\":\"string\",\"description\":\"Delivery target\"},\"threadId\":{\"anyOf\":[{\"type\":\"string\"},{\"type\":\"number\"}],\"description\":\"Thread/topic id\"},\"bestEffort\":{\"type\":\"boolean\"},\"accountId\":{\"type\":\"string\",\"description\":\"Delivery account\"},\"failureDestination\":{\"type\":\"object\",\"properties\":{\"channel\":{\"type\":\"string\"},\"to\":{\"type\":\"string\"},\"accountId\":{\"type\":\"string\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]}},\"additionalProperties\":true}},\"additionalProperties\":true},\"agentId\":{\"type\":\"string\",\"description\":\"Agent id, or null to keep it unset\"},\"description\":{\"type\":\"string\",\"description\":\"Human description\"},\"enabled\":{\"type\":\"boolean\"},\"deleteAfterRun\":{\"type\":\"boolean\",\"description\":\"Delete after first run\"},\"sessionKey\":{\"type\":\"string\",\"description\":\"Explicit session key, or null to clear it\"},\"failureAlert\":{\"type\":\"object\",\"properties\":{\"after\":{\"type\":\"number\",\"description\":\"Failures before alert\"},\"channel\":{\"type\":\"string\",\"description\":\"Alert channel\"},\"to\":{\"type\":\"string\",\"description\":\"Alert target\"},\"cooldownMs\":{\"type\":\"number\",\"description\":\"Alert cooldown ms\"},\"includeSkipped\":{\"type\":\"boolean\",\"description\":\"Skipped runs count toward alert\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]},\"accountId\":{\"type\":\"string\"}},\"additionalProperties\":true,\"description\":\"Failure alert object; false disables alerts\"}},\"additionalProperties\":true},\"jobId\":{\"type\":\"string\"},\"id\":{\"type\":\"string\"},\"patch\":{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\",\"description\":\"Job name\"},\"schedule\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"at\",\"every\",\"cron\"],\"description\":\"Schedule kind\"},\"at\":{\"type\":\"string\",\"description\":\"ISO-8601 time (kind=at)\"},\"everyMs\":{\"type\":\"number\",\"description\":\"Interval ms (kind=every)\"},\"anchorMs\":{\"type\":\"number\",\"description\":\"Start anchor ms (kind=every)\"},\"expr\":{\"type\":\"string\",\"description\":\"Cron expr in tz wall-clock time; do not convert to UTC. Omitted tz => Gateway host local timezone. Example 6pm Shanghai daily: expr \\\"0 18 * * *\\\", tz \\\"Asia/Shanghai\\\".\"},\"tz\":{\"type\":\"string\",\"description\":\"IANA timezone for cron wall-clock fields, e.g. \\\"Asia/Shanghai\\\"; omitted => Gateway host local timezone.\"},\"staggerMs\":{\"type\":\"number\",\"description\":\"Jitter ms (kind=cron)\"}},\"additionalProperties\":true},\"sessionTarget\":{\"type\":\"string\",\"description\":\"Session target\"},\"wakeMode\":{\"type\":\"string\",\"enum\":[\"now\",\"next-heartbeat\"]},\"payload\":{\"type\":\"object\",\"properties\":{\"kind\":{\"type\":\"string\",\"enum\":[\"systemEvent\",\"agentTurn\"],\"description\":\"Payload kind\"},\"text\":{\"type\":\"string\",\"description\":\"systemEvent text\"},\"message\":{\"type\":\"string\",\"description\":\"agentTurn prompt\"},\"model\":{\"type\":\"string\",\"description\":\"Model override\"},\"thinking\":{\"type\":\"string\",\"description\":\"Thinking override\"},\"timeoutSeconds\":{\"type\":\"number\"},\"lightContext\":{\"type\":\"boolean\"},\"allowUnsafeExternalContent\":{\"type\":\"boolean\"},\"fallbacks\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Fallback models\"},\"toolsAllow\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Allowed tool ids, or null to clear\"}},\"additionalProperties\":true},\"delivery\":{\"type\":\"object\",\"properties\":{\"mode\":{\"type\":\"string\",\"enum\":[\"none\",\"announce\",\"webhook\"],\"description\":\"Delivery mode\"},\"channel\":{\"type\":\"string\",\"description\":\"Delivery channel\"},\"to\":{\"type\":\"string\",\"description\":\"Delivery target\"},\"threadId\":{\"anyOf\":[{\"type\":\"string\"},{\"type\":\"number\"}],\"description\":\"Thread/topic id\"},\"bestEffort\":{\"type\":\"boolean\"},\"accountId\":{\"type\":\"string\",\"description\":\"Delivery account\"},\"failureDestination\":{\"type\":\"object\",\"properties\":{\"channel\":{\"type\":\"string\"},\"to\":{\"type\":\"string\"},\"accountId\":{\"type\":\"string\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]}},\"additionalProperties\":true}},\"additionalProperties\":true},\"description\":{\"type\":\"string\"},\"enabled\":{\"type\":\"boolean\"},\"deleteAfterRun\":{\"type\":\"boolean\"},\"agentId\":{\"type\":\"string\",\"description\":\"Agent id, or null to clear it\"},\"sessionKey\":{\"type\":\"string\",\"description\":\"Explicit session key, or null to clear it\"},\"failureAlert\":{\"type\":\"object\",\"properties\":{\"after\":{\"type\":\"number\",\"description\":\"Failures before alert\"},\"channel\":{\"type\":\"string\",\"description\":\"Alert channel\"},\"to\":{\"type\":\"string\",\"description\":\"Alert target\"},\"cooldownMs\":{\"type\":\"number\",\"description\":\"Alert cooldown ms\"},\"includeSkipped\":{\"type\":\"boolean\",\"description\":\"Skipped runs count toward alert\"},\"mode\":{\"type\":\"string\",\"enum\":[\"announce\",\"webhook\"]},\"accountId\":{\"type\":\"string\"}},\"additionalProperties\":true,\"description\":\"Failure alert object; false disables alerts\"}},\"additionalProperties\":true},\"text\":{\"type\":\"string\"},\"mode\":{\"type\":\"string\",\"enum\":[\"now\",\"next-heartbeat\"]},\"runMode\":{\"type\":\"string\",\"enum\":[\"due\",\"force\"]},\"contextMessages\":{\"type\":\"number\",\"minimum\":0,\"maximum\":10},\"agentId\":{\"type\":\"string\",\"description\":\"List filter: agent id\"}},\"additionalProperties\":true}}},{\"type\":\"function\",\"function\":{\"name\":\"dir_fetch\",\"description\":\"Retrieve a directory tree from a paired node as a gzipped tarball, unpack it on the gateway, and return a manifest of saved paths. Use to pull source trees, asset folders, or log directories in a single round-trip. The unpacked files live on the GATEWAY (not your local machine); pass localPath into other tools or use file_fetch on individual entries to ship them elsewhere. Rejects trees larger than 16 MB compressed. Requires operator opt-in: gateway.nodes.allowCommands must include 'dir.fetch' AND plugins.entries.file-transfer.config.nodes.<node>.allowReadPaths must match the directory path.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id, name, or IP. Resolves the same way as the nodes tool.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path to the directory on the node to fetch. Canonicalized server-side.\"},\"maxBytes\":{\"type\":\"number\",\"description\":\"Max gzipped tarball bytes to fetch. Default 8 MB, hard ceiling 16 MB (single round-trip).\"},\"includeDotfiles\":{\"type\":\"boolean\",\"description\":\"Reserved for v2; currently always includes dotfiles (v1 quirk in BSD tar).\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"dir_list\",\"description\":\"Retrieve a structured directory listing from a paired node. Returns file and subdirectory metadata (name, path, size, mimeType, isDir, mtime) without transferring file content. Use this to discover what files exist before fetching them with file_fetch. Pagination is offset-based; pass nextPageToken from the previous result. Requires operator opt-in: gateway.nodes.allowCommands must include 'dir.list' AND plugins.entries.file-transfer.config.nodes.<node>.allowReadPaths must match the directory path. Without policy configured, every call is denied.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id, name, or IP. Resolves the same way as the nodes tool.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path to the directory on the node. Canonicalized server-side.\"},\"pageToken\":{\"type\":\"string\",\"description\":\"Pagination token from a previous dir_list call. Omit to start from the beginning.\"},\"maxEntries\":{\"type\":\"number\",\"description\":\"Max entries per page. Default 200, hard ceiling 5000.\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"edit\",\"description\":\"Edit a single file using exact text replacement. Every edits[].oldText must match a unique, non-overlapping region of the original file. If two changes affect the same block or nearby lines, merge them into one edit instead of emitting overlapping edits. Do not include large unchanged regions just to connect distant changes.\",\"parameters\":{\"type\":\"object\",\"required\":[\"path\",\"edits\"],\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Path to the file to edit (relative or absolute)\"},\"edits\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"oldText\",\"newText\"],\"properties\":{\"oldText\":{\"type\":\"string\",\"description\":\"Exact text for one targeted replacement. It must be unique in the original file and must not overlap with any other edits[].oldText in the same call.\"},\"newText\":{\"type\":\"string\",\"description\":\"Replacement text for this targeted edit.\"}},\"additionalProperties\":false},\"description\":\"One or more targeted replacements. Each edit is matched against the original file, not incrementally. Do not include overlapping or nested edits. If two changes touch the same block or nearby lines, merge them into one edit instead.\"}},\"additionalProperties\":false}}},{\"type\":\"function\",\"function\":{\"name\":\"exec\",\"description\":\"Execute shell commands with background continuation for work that starts now. Use yieldMs/background to continue later via process tool. For long-running work started now, rely on automatic completion wake when it is enabled and the command emits output or fails; otherwise use process to confirm completion. Use process whenever you need logs, status, input, or intervention. Do not use exec sleep or delay loops for reminders or deferred follow-ups; use cron instead. Use pty=true for TTY-required commands (terminal UIs, coding agents).\",\"parameters\":{\"type\":\"object\",\"required\":[\"command\"],\"properties\":{\"command\":{\"type\":\"string\",\"description\":\"Shell command to execute\"},\"workdir\":{\"type\":\"string\",\"description\":\"Working directory (defaults to cwd)\"},\"env\":{\"type\":\"object\",\"patternProperties\":{\"^.*$\":{\"type\":\"string\"}}},\"yieldMs\":{\"type\":\"number\",\"description\":\"Milliseconds to wait before backgrounding (default 10000)\"},\"background\":{\"type\":\"boolean\",\"description\":\"Run in background immediately\"},\"timeout\":{\"type\":\"number\",\"description\":\"Timeout in seconds (optional, kills process on expiry)\"},\"pty\":{\"type\":\"boolean\",\"description\":\"Run in a pseudo-terminal (PTY) when available (TTY-required CLIs, coding agents)\"},\"elevated\":{\"type\":\"boolean\",\"description\":\"Run on the host with elevated permissions (if allowed)\"},\"host\":{\"type\":\"string\",\"enum\":[\"auto\",\"sandbox\",\"gateway\",\"node\"],\"description\":\"Exec host/target (auto|sandbox|gateway|node).\"},\"security\":{\"type\":\"string\",\"description\":\"Ignored for normal calls; exec security is set by tools.exec.security and host approvals.\"},\"ask\":{\"type\":\"string\",\"description\":\"Exec ask mode (off|on-miss|always).\"},\"node\":{\"type\":\"string\",\"description\":\"Node id/name for host=node.\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"file_fetch\",\"description\":\"Retrieve a file from a paired node by absolute path. Returns image content blocks for image MIME types, inlines small text files (\u22648 KB) as text content, and saves everything else under the gateway media store with a path you can pass to file_write or other tools. Use this for screenshots, photos, receipts, logs, source files. Pair with file_write to copy a file from one node to another (no exec/cp shell-out needed). Requires operator opt-in: gateway.nodes.allowCommands must include 'file.fetch' AND plugins.entries.file-transfer.config.nodes.<node>.allowReadPaths must match the path. Without policy configured, every call is denied.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id, name, or IP. Resolves the same way as the nodes tool.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path to the file on the node. Canonicalized server-side.\"},\"maxBytes\":{\"type\":\"number\",\"description\":\"Max bytes to fetch. Default 8 MB, hard ceiling 16 MB (single round-trip).\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"file_write\",\"description\":\"Write file bytes to a paired node by absolute path. Atomic write (temp + rename). Refuses to overwrite by default \u2014 pass overwrite=true to replace. Refuses to write through symlink targets unless policy explicitly allows following symlinks. Pair with file_fetch by passing its mediaId as sourceMediaId for binary copy. Requires operator opt-in: gateway.nodes.allowCommands must include 'file.write' AND plugins.entries.file-transfer.config.nodes.<node>.allowWritePaths must match the destination path. Without policy configured, every call is denied.\",\"parameters\":{\"type\":\"object\",\"required\":[\"node\",\"path\"],\"properties\":{\"node\":{\"type\":\"string\",\"description\":\"Node id or display name to write the file on.\"},\"path\":{\"type\":\"string\",\"description\":\"Absolute path on the node to write. Canonicalized server-side.\"},\"contentBase64\":{\"type\":\"string\",\"description\":\"Base64-encoded bytes to write. Maximum 16 MB after decode.\"},\"sourceMediaId\":{\"type\":\"string\",\"description\":\"Media id returned by file_fetch. Preferred for binary copies because bytes stay in the gateway media store.\"},\"mimeType\":{\"type\":\"string\",\"description\":\"Content type hint. Not validated against the content.\"},\"overwrite\":{\"type\":\"boolean\",\"description\":\"Allow overwriting an existing file. Default false.\",\"default\":false},\"createParents\":{\"type\":\"boolean\",\"description\":\"Create missing parent directories (mkdir -p). Default false.\",\"default\":false}}}}},{\"type\":\"function\",\"function\":{\"name\":\"gateway\",\"description\":\"Gateway restart/config/update. Before config edits, use config.schema.lookup with targeted dot path. Prefer config.patch for partial merge; config.apply only full replace. Writes hot-reload or restart as needed. Always pass human `note` for post-restart delivery. If still owe the user a reply, pass one-shot `continuationMessage`; do not write restart sentinel files directly.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"restart\",\"config.get\",\"config.schema.lookup\",\"config.apply\",\"config.patch\",\"update.run\"]},\"delayMs\":{\"type\":\"number\"},\"reason\":{\"type\":\"string\"},\"continuationMessage\":{\"type\":\"string\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"path\":{\"type\":\"string\"},\"raw\":{\"type\":\"string\"},\"baseHash\":{\"type\":\"string\"},\"sessionKey\":{\"type\":\"string\"},\"note\":{\"type\":\"string\"},\"restartDelayMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"image\",\"description\":\"Analyze images with available vision model. Use image for one path/URL, images for max 20. Prompt says what to inspect.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"prompt\":{\"type\":\"string\"},\"image\":{\"type\":\"string\",\"description\":\"One image path/URL.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Image paths/URLs; maxImages default 20.\"},\"model\":{\"type\":\"string\"},\"maxBytesMb\":{\"type\":\"number\"},\"maxImages\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"image_generate\",\"description\":\"Create/edit images. Session chats: background task; do not call image_generate again for same request; wait completion, then send attachments via message tool. Transparent: outputFormat=\\\"png\\\" or \\\"webp\\\" + background=\\\"transparent\\\"; OpenAI also supports openai.background and routes default model to gpt-image-1.5. Use action=\\\"list\\\" for providers/models/readiness/auth, \\\"status\\\" for active task.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"\\\"generate\\\" default, \\\"status\\\" active task, \\\"list\\\" providers/models.\"},\"prompt\":{\"type\":\"string\",\"description\":\"Image prompt.\"},\"image\":{\"type\":\"string\",\"description\":\"Reference image path/URL for edit.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference images for edit; max 5.\"},\"model\":{\"type\":\"string\",\"description\":\"Provider/model override, e.g. openai/gpt-image-2; transparent OpenAI: openai/gpt-image-1.5.\"},\"filename\":{\"type\":\"string\",\"description\":\"Output filename hint; basename preserved in managed media dir.\"},\"size\":{\"type\":\"string\",\"description\":\"Size hint: 1024x1024, 1536x1024, 1024x1536, 2048x2048, 3840x2160.\"},\"aspectRatio\":{\"type\":\"string\",\"description\":\"Aspect ratio: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.\"},\"resolution\":{\"type\":\"string\",\"description\":\"Resolution: 1K, 2K, 4K; useful for Google.\"},\"quality\":{\"type\":\"string\",\"enum\":[\"low\",\"medium\",\"high\",\"auto\"],\"description\":\"Quality: low, medium, high, auto.\"},\"outputFormat\":{\"type\":\"string\",\"enum\":[\"png\",\"jpeg\",\"webp\"],\"description\":\"Output format: png, jpeg, webp.\"},\"background\":{\"type\":\"string\",\"enum\":[\"transparent\",\"opaque\",\"auto\"],\"description\":\"Background: transparent, opaque, auto. Transparent needs png/webp output.\"},\"openai\":{\"type\":\"object\",\"properties\":{\"background\":{\"type\":\"string\",\"enum\":[\"transparent\",\"opaque\",\"auto\"],\"description\":\"OpenAI background: transparent, opaque, auto. Transparent needs png/webp; default model routes to gpt-image-1.5.\"},\"moderation\":{\"type\":\"string\",\"enum\":[\"low\",\"auto\"],\"description\":\"OpenAI moderation: low, auto.\"},\"outputCompression\":{\"type\":\"number\",\"description\":\"OpenAI jpeg/webp compression 0-100.\",\"minimum\":0,\"maximum\":100},\"user\":{\"type\":\"string\",\"description\":\"OpenAI stable end-user id.\"}}},\"count\":{\"type\":\"number\",\"description\":\"Image count 1-4.\",\"minimum\":1,\"maximum\":4},\"timeoutMs\":{\"type\":\"number\",\"description\":\"Provider timeout ms.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"memory_get\",\"description\":\"Safe exact excerpt read from MEMORY.md or memory/*.md. Defaults to a bounded excerpt when lines are omitted, includes truncation/continuation info when more content exists, and `corpus=wiki` reads from registered compiled-wiki supplements.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"path\":{\"type\":\"string\"},\"from\":{\"type\":\"number\"},\"lines\":{\"type\":\"number\"},\"corpus\":{\"type\":\"string\",\"enum\":[\"memory\",\"wiki\",\"all\"]}},\"required\":[\"path\"],\"additionalProperties\":false}}},{\"type\":\"function\",\"function\":{\"name\":\"memory_search\",\"description\":\"Mandatory recall step: semantically search MEMORY.md + memory/*.md (and optional session transcripts) before answering questions about prior work, decisions, dates, people, preferences, or todos. Optional `corpus=wiki` or `corpus=all` also searches registered compiled-wiki supplements. `corpus=memory` restricts hits to indexed memory files (excludes session transcript chunks from ranking). `corpus=sessions` restricts hits to indexed session transcripts (same visibility rules as session history tools). If response has disabled=true, memory retrieval is unavailable and should be surfaced to the user.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"},\"maxResults\":{\"type\":\"number\"},\"minScore\":{\"type\":\"number\"},\"corpus\":{\"type\":\"string\",\"enum\":[\"memory\",\"wiki\",\"all\",\"sessions\"]}},\"required\":[\"query\"],\"additionalProperties\":false}}},{\"type\":\"function\",\"function\":{\"name\":\"message\",\"description\":\"Send/delete/manage channel messages. Supports actions: delete, edit, poll, react, send, topic-create, topic-edit. This turn: use action=\\\"send\\\" with message for visible replies to the current source conversation. target defaults to the current source conversation; omit unless sending elsewhere. Normal final answers stay private.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"send\",\"poll\",\"react\",\"delete\",\"edit\",\"topic-create\",\"topic-edit\"]},\"channel\":{\"type\":\"string\"},\"target\":{\"type\":\"string\",\"description\":\"Recipient/channel: E.164 for WhatsApp/Signal, Telegram chat id/@username, Discord/Slack/Mattermost <channelId|user:ID|channel:ID>, or iMessage handle/chat_id\"},\"targets\":{\"type\":\"array\",\"items\":{\"type\":\"string\",\"description\":\"Recipient/channel targets (same format as --target); accepts ids or names when the directory is available.\"}},\"accountId\":{\"type\":\"string\"},\"dryRun\":{\"type\":\"boolean\"},\"message\":{\"type\":\"string\"},\"effectId\":{\"type\":\"string\",\"description\":\"Effect id/name for sendWithEffect.\"},\"effect\":{\"type\":\"string\",\"description\":\"Alias for effectId.\"},\"media\":{\"type\":\"string\",\"description\":\"Media URL/path. data: use buffer.\"},\"filename\":{\"type\":\"string\"},\"buffer\":{\"type\":\"string\",\"description\":\"Base64 attachment payload; data URL ok.\"},\"contentType\":{\"type\":\"string\"},\"mimeType\":{\"type\":\"string\"},\"caption\":{\"type\":\"string\"},\"path\":{\"type\":\"string\"},\"filePath\":{\"type\":\"string\"},\"attachments\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{\"type\":{\"type\":\"string\",\"enum\":[\"image\",\"audio\",\"video\",\"file\"]},\"media\":{\"type\":\"string\"},\"mediaUrl\":{\"type\":\"string\"},\"path\":{\"type\":\"string\"},\"filePath\":{\"type\":\"string\"},\"fileUrl\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"name\":{\"type\":\"string\"},\"mimeType\":{\"type\":\"string\"}}},\"description\":\"Structured attachments; each needs media/mediaUrl/path/filePath/fileUrl/url.\"},\"replyTo\":{\"type\":\"string\"},\"threadId\":{\"type\":\"string\"},\"asVoice\":{\"type\":\"boolean\"},\"silent\":{\"type\":\"boolean\"},\"quoteText\":{\"type\":\"string\",\"description\":\"Telegram reply quote text.\"},\"bestEffort\":{\"type\":\"boolean\"},\"gifPlayback\":{\"type\":\"boolean\"},\"forceDocument\":{\"type\":\"boolean\",\"description\":\"Send image/GIF/video as document; avoids compression.\"},\"asDocument\":{\"type\":\"boolean\",\"description\":\"Alias for forceDocument.\"},\"presentation\":{\"type\":\"object\",\"required\":[\"blocks\"],\"properties\":{\"title\":{\"type\":\"string\"},\"tone\":{\"type\":\"string\",\"enum\":[\"info\",\"success\",\"warning\",\"danger\",\"neutral\"]},\"blocks\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"type\"],\"properties\":{\"type\":{\"type\":\"string\",\"enum\":[\"text\",\"context\",\"divider\",\"buttons\",\"select\"]},\"text\":{\"type\":\"string\"},\"buttons\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"label\"],\"properties\":{\"label\":{\"type\":\"string\"},\"value\":{\"type\":\"string\"},\"url\":{\"type\":\"string\"},\"webApp\":{\"type\":\"object\",\"required\":[\"url\"],\"properties\":{\"url\":{\"type\":\"string\"}}},\"web_app\":{\"type\":\"object\",\"required\":[\"url\"],\"properties\":{\"url\":{\"type\":\"string\"}}},\"disabled\":{\"type\":\"boolean\"},\"reusable\":{\"type\":\"boolean\"},\"style\":{\"type\":\"string\",\"enum\":[\"primary\",\"secondary\",\"success\",\"danger\"]}}}},\"placeholder\":{\"type\":\"string\"},\"options\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"label\",\"value\"],\"properties\":{\"label\":{\"type\":\"string\"},\"value\":{\"type\":\"string\"}}}}}}}},\"description\":\"Rich message payload: text/buttons/selects/context. Unsupported blocks degrade to text.\"},\"delivery\":{\"type\":\"object\",\"properties\":{\"pin\":{\"anyOf\":[{\"type\":\"boolean\"},{\"type\":\"object\",\"required\":[\"enabled\"],\"properties\":{\"enabled\":{\"type\":\"boolean\"},\"notify\":{\"type\":\"boolean\"},\"required\":{\"type\":\"boolean\"}}}]}},\"description\":\"Delivery prefs. pin requests pin when channel supports it.\"},\"messageId\":{\"type\":\"string\",\"description\":\"Target message id for read/react/edit/delete/pin/unpin. Reaction-like defaults current inbound id when available.\"},\"message_id\":{\"type\":\"string\",\"description\":\"snake_case alias of messageId; same defaults.\"},\"emoji\":{\"type\":\"string\"},\"remove\":{\"type\":\"boolean\"},\"trackToolCalls\":{\"type\":\"boolean\",\"description\":\"For current-message reaction, make reacted message the tool-progress reaction target.\"},\"track_tool_calls\":{\"type\":\"boolean\",\"description\":\"snake_case alias of trackToolCalls.\"},\"targetAuthor\":{\"type\":\"string\"},\"targetAuthorUuid\":{\"type\":\"string\"},\"groupId\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\"},\"pageSize\":{\"type\":\"number\"},\"pageToken\":{\"type\":\"string\"},\"before\":{\"type\":\"string\"},\"after\":{\"type\":\"string\"},\"around\":{\"type\":\"string\"},\"fromMe\":{\"type\":\"boolean\"},\"includeArchived\":{\"type\":\"boolean\"},\"pollId\":{\"type\":\"string\"},\"pollOptionId\":{\"type\":\"string\",\"description\":\"Poll answer id.\"},\"pollOptionIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\",\"description\":\"Poll answer ids for multiselect.\"}},\"pollOptionIndex\":{\"type\":\"number\",\"description\":\"1-based poll option number.\"},\"pollOptionIndexes\":{\"type\":\"array\",\"items\":{\"type\":\"number\",\"description\":\"1-based poll option numbers for multiselect.\"}},\"pollQuestion\":{\"type\":\"string\"},\"pollOption\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"pollDurationHours\":{\"type\":\"number\"},\"pollMulti\":{\"type\":\"boolean\"},\"channelId\":{\"type\":\"string\",\"description\":\"Channel id filter.\"},\"chatId\":{\"type\":\"string\",\"description\":\"Chat id for chat metadata.\"},\"channelIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\",\"description\":\"Channel id filter.\"}},\"memberId\":{\"type\":\"string\"},\"memberIdType\":{\"type\":\"string\"},\"guildId\":{\"type\":\"string\"},\"userId\":{\"type\":\"string\"},\"openId\":{\"type\":\"string\"},\"unionId\":{\"type\":\"string\"},\"authorId\":{\"type\":\"string\"},\"authorIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"roleId\":{\"type\":\"string\"},\"roleIds\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"participant\":{\"type\":\"string\"},\"includeMembers\":{\"type\":\"boolean\"},\"members\":{\"type\":\"boolean\"},\"scope\":{\"type\":\"string\"},\"kind\":{\"type\":\"string\"},\"fileId\":{\"type\":\"string\"},\"emojiName\":{\"type\":\"string\"},\"stickerId\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"stickerName\":{\"type\":\"string\"},\"stickerDesc\":{\"type\":\"string\"},\"stickerTags\":{\"type\":\"string\"},\"threadName\":{\"type\":\"string\"},\"autoArchiveMin\":{\"type\":\"number\"},\"appliedTags\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"query\":{\"type\":\"string\"},\"eventName\":{\"type\":\"string\"},\"eventType\":{\"type\":\"string\"},\"startTime\":{\"type\":\"string\"},\"endTime\":{\"type\":\"string\"},\"desc\":{\"type\":\"string\"},\"location\":{\"type\":\"string\"},\"image\":{\"type\":\"string\",\"description\":\"Event cover image URL/path.\"},\"durationMin\":{\"type\":\"number\"},\"until\":{\"type\":\"string\"},\"reason\":{\"type\":\"string\"},\"deleteDays\":{\"type\":\"number\"},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"name\":{\"type\":\"string\"},\"channelType\":{\"type\":\"number\",\"description\":\"Numeric channel type, e.g. Discord. Avoids JSON Schema `type` collision.\"},\"parentId\":{\"type\":\"string\"},\"topic\":{\"type\":\"string\"},\"position\":{\"type\":\"number\"},\"nsfw\":{\"type\":\"boolean\"},\"rateLimitPerUser\":{\"type\":\"number\"},\"categoryId\":{\"type\":\"string\"},\"clearParent\":{\"type\":\"boolean\",\"description\":\"Clear parent/category when supported.\"},\"activityType\":{\"type\":\"string\",\"description\":\"Activity type: playing, streaming, listening, watching, competing, custom.\"},\"activityName\":{\"type\":\"string\",\"description\":\"Activity name shown in sidebar; ignored for custom.\"},\"activityUrl\":{\"type\":\"string\",\"description\":\"Streaming URL; streaming type only.\"},\"activityState\":{\"type\":\"string\",\"description\":\"State text; custom type uses as status text.\"},\"status\":{\"type\":\"string\",\"description\":\"Bot status: online, dnd, idle, invisible.\"},\"pollDurationSeconds\":{\"type\":\"number\"},\"pollAnonymous\":{\"type\":\"boolean\"},\"pollPublic\":{\"type\":\"boolean\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"music_generate\",\"description\":\"Create audio/music for song, jingle, beat, loop, soundtrack, anthem, instrumental requests. If user asks make/generate/create song/music, call music_generate; do not just write lyrics unless lyrics/text only. Prompt gets style/genre/mood/tempo/instruments/purpose. lyrics only exact sung words. Session chats: background task; do not call again for same request; wait completion, send attachments via message tool. \\\"status\\\" checks active task.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"\\\"generate\\\" default, \\\"status\\\" active task, \\\"list\\\" providers/models.\"},\"prompt\":{\"type\":\"string\",\"description\":\"Music prompt: style, genre, mood, purpose.\"},\"lyrics\":{\"type\":\"string\",\"description\":\"Exact sung lyrics only when the user supplies lyrics or asks for vocal words. For song/style requests, use prompt instead.\"},\"instrumental\":{\"type\":\"boolean\",\"description\":\"Instrumental-only toggle.\"},\"image\":{\"type\":\"string\",\"description\":\"Reference image path/URL.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference images; max 10.\"},\"model\":{\"type\":\"string\",\"description\":\"Provider/model override, e.g. google/lyria-3-pro-preview.\"},\"durationSeconds\":{\"type\":\"number\",\"description\":\"Target seconds; provider may clamp.\",\"minimum\":1},\"format\":{\"type\":\"string\",\"description\":\"Output format: mp3, wav.\"},\"filename\":{\"type\":\"string\",\"description\":\"Output filename hint; basename preserved in managed media dir.\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"nodes\",\"description\":\"Discover/control paired nodes: status, describe, pairing, notify, camera/photos/screen/location/notifications/invoke. Use file_fetch for files.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"status\",\"describe\",\"pending\",\"approve\",\"reject\",\"notify\",\"camera_snap\",\"camera_list\",\"camera_clip\",\"photos_latest\",\"screen_record\",\"location_get\",\"notifications_list\",\"notifications_action\",\"device_status\",\"device_info\",\"device_permissions\",\"device_health\",\"invoke\"]},\"gatewayUrl\":{\"type\":\"string\"},\"gatewayToken\":{\"type\":\"string\"},\"timeoutMs\":{\"type\":\"number\"},\"node\":{\"type\":\"string\"},\"requestId\":{\"type\":\"string\"},\"title\":{\"type\":\"string\"},\"body\":{\"type\":\"string\"},\"sound\":{\"type\":\"string\"},\"priority\":{\"type\":\"string\",\"enum\":[\"passive\",\"active\",\"timeSensitive\"]},\"delivery\":{\"type\":\"string\",\"enum\":[\"system\",\"overlay\",\"auto\"]},\"facing\":{\"type\":\"string\",\"enum\":[\"front\",\"back\",\"both\"],\"description\":\"camera_snap: front/back/both; camera_clip: front/back only.\"},\"maxWidth\":{\"type\":\"number\"},\"quality\":{\"type\":\"number\"},\"delayMs\":{\"type\":\"number\"},\"deviceId\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\"},\"duration\":{\"type\":\"string\"},\"durationMs\":{\"type\":\"number\",\"maximum\":300000},\"includeAudio\":{\"type\":\"boolean\"},\"fps\":{\"type\":\"number\"},\"screenIndex\":{\"type\":\"number\"},\"outPath\":{\"type\":\"string\"},\"maxAgeMs\":{\"type\":\"number\"},\"locationTimeoutMs\":{\"type\":\"number\"},\"desiredAccuracy\":{\"type\":\"string\",\"enum\":[\"coarse\",\"balanced\",\"precise\"]},\"notificationAction\":{\"type\":\"string\",\"enum\":[\"open\",\"dismiss\",\"reply\"]},\"notificationKey\":{\"type\":\"string\"},\"notificationReplyText\":{\"type\":\"string\"},\"invokeCommand\":{\"type\":\"string\"},\"invokeParamsJson\":{\"type\":\"string\"},\"invokeTimeoutMs\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"pdf\",\"description\":\"Analyze PDFs with model. Anthropic/Google native PDF when supported; else text/image extraction. Use pdf for one, pdfs for max 10; prompt says what to inspect.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"prompt\":{\"type\":\"string\"},\"pdf\":{\"type\":\"string\",\"description\":\"One PDF path/URL.\"},\"pdfs\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"PDF paths/URLs; max 10.\"},\"pages\":{\"type\":\"string\",\"description\":\"Pages, e.g. \\\"1-5\\\", \\\"1,3,5-7\\\"; default all.\"},\"model\":{\"type\":\"string\"},\"maxBytesMb\":{\"type\":\"number\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"process\",\"description\":\"Manage running exec sessions for commands already started: list, poll, log, write, send-keys, submit, paste, kill. Use poll/log when you need status, logs, quiet-success confirmation, or completion confirmation when automatic completion wake is unavailable. Use poll/log also for input-wait hints. Use write/send-keys/submit/paste/kill for input or intervention. Do not use process polling to emulate timers or reminders; use cron for scheduled follow-ups.\",\"parameters\":{\"type\":\"object\",\"required\":[\"action\"],\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"Process action (list|poll|log|write|send-keys|submit|paste|kill|clear|remove)\"},\"sessionId\":{\"type\":\"string\",\"description\":\"Session id for actions other than list\"},\"data\":{\"type\":\"string\",\"description\":\"Data to write for write\"},\"keys\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Key tokens to send for send-keys\"},\"hex\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Hex bytes to send for send-keys\"},\"literal\":{\"type\":\"string\",\"description\":\"Literal string for send-keys\"},\"text\":{\"type\":\"string\",\"description\":\"Text to paste for paste\"},\"bracketed\":{\"type\":\"boolean\",\"description\":\"Wrap paste in bracketed mode\"},\"eof\":{\"type\":\"boolean\",\"description\":\"Close stdin after write\"},\"offset\":{\"type\":\"number\",\"description\":\"Log offset\"},\"limit\":{\"type\":\"number\",\"description\":\"Log length\"},\"timeout\":{\"type\":\"number\",\"description\":\"For poll: wait up to this many milliseconds before returning; max 30000 ms, higher values are clamped to 30000\",\"minimum\":0}}}}},{\"type\":\"function\",\"function\":{\"name\":\"read\",\"description\":\"Read the contents of a file. Supports text files and images (jpg, png, gif, webp). Images are sent as attachments. For text files, output is truncated to 2000 lines or 50KB (whichever is hit first). Use offset/limit for large files. When you need the full file, continue with offset until complete.\",\"parameters\":{\"type\":\"object\",\"required\":[\"path\"],\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Path to the file to read (relative or absolute)\"},\"offset\":{\"type\":\"number\",\"description\":\"Line number to start reading from (1-indexed)\"},\"limit\":{\"type\":\"number\",\"description\":\"Maximum number of lines to read\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"session_status\",\"description\":\"Show /status-like card for current/visible session: model, usage, time, cost, tasks. Use `sessionKey=\\\"current\\\"` for current session; UI labels like `openclaw-tui` are not keys. `model` sets session override; `model=default` resets. Use for active model/session config questions.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"sessionKey\":{\"type\":\"string\"},\"model\":{\"type\":\"string\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_history\",\"description\":\"Fetch sanitized history for visible session. Use before replying, debugging, resuming; supports limits/tool messages.\",\"parameters\":{\"type\":\"object\",\"required\":[\"sessionKey\"],\"properties\":{\"sessionKey\":{\"type\":\"string\"},\"limit\":{\"type\":\"number\",\"minimum\":1},\"includeTools\":{\"type\":\"boolean\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_list\",\"description\":\"List visible sessions; filter by kind, label, agentId, search, activity. Use before sessions_history or sessions_send target selection.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"kinds\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"limit\":{\"type\":\"number\",\"minimum\":1},\"activeMinutes\":{\"type\":\"number\",\"minimum\":1},\"messageLimit\":{\"type\":\"number\",\"minimum\":0},\"label\":{\"type\":\"string\",\"minLength\":1},\"agentId\":{\"type\":\"string\",\"minLength\":1,\"maxLength\":64},\"search\":{\"type\":\"string\",\"minLength\":1},\"includeDerivedTitles\":{\"type\":\"boolean\"},\"includeLastMessage\":{\"type\":\"boolean\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_send\",\"description\":\"Send message to visible session by sessionKey/label, or configured agent by agentId. Thread-scoped chats rejected; target parent channel session. Creates missing configured-agent main session; waits for reply when available.\",\"parameters\":{\"type\":\"object\",\"required\":[\"message\"],\"properties\":{\"sessionKey\":{\"type\":\"string\"},\"label\":{\"type\":\"string\",\"minLength\":1,\"maxLength\":512},\"agentId\":{\"type\":\"string\",\"minLength\":1,\"maxLength\":64},\"message\":{\"type\":\"string\"},\"timeoutSeconds\":{\"type\":\"number\",\"minimum\":0}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_spawn\",\"description\":\"Spawn clean child session; default `runtime=\\\"subagent\\\"`. `mode=\\\"run\\\"` one-shot background work. Subagents inherit parent workspace. Native subagents get task in first visible `[Subagent Task]` message. Native only: `context=\\\"fork\\\"` only when child needs current transcript; else omit or `isolated`. Use for fresh child-session work.\",\"parameters\":{\"type\":\"object\",\"required\":[\"task\"],\"properties\":{\"task\":{\"type\":\"string\"},\"taskName\":{\"type\":\"string\",\"description\":\"Stable alias for later targeting; lowercase letters/digits/underscores, starts letter.\"},\"label\":{\"type\":\"string\"},\"runtime\":{\"type\":\"string\",\"enum\":[\"subagent\"]},\"agentId\":{\"type\":\"string\"},\"model\":{\"type\":\"string\"},\"thinking\":{\"type\":\"string\"},\"cwd\":{\"type\":\"string\"},\"runTimeoutSeconds\":{\"type\":\"number\",\"minimum\":0},\"timeoutSeconds\":{\"type\":\"number\",\"minimum\":0},\"mode\":{\"type\":\"string\",\"enum\":[\"run\"]},\"cleanup\":{\"type\":\"string\",\"enum\":[\"delete\",\"keep\"]},\"sandbox\":{\"type\":\"string\",\"enum\":[\"inherit\",\"require\"]},\"context\":{\"type\":\"string\",\"enum\":[\"isolated\",\"fork\"],\"description\":\"Native context. Omit/\\\"isolated\\\" for clean child; \\\"fork\\\" only when child needs requester transcript.\"},\"lightContext\":{\"type\":\"boolean\",\"description\":\"Light bootstrap context; runtime=\\\"subagent\\\" only.\"},\"attachments\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"name\",\"content\"],\"properties\":{\"name\":{\"type\":\"string\"},\"content\":{\"type\":\"string\"},\"encoding\":{\"type\":\"string\",\"enum\":[\"utf8\",\"base64\"]},\"mimeType\":{\"type\":\"string\"}}},\"maxItems\":50},\"attachAs\":{\"type\":\"object\",\"properties\":{\"mountPath\":{\"type\":\"string\"}}}}}}},{\"type\":\"function\",\"function\":{\"name\":\"sessions_yield\",\"description\":\"End current turn. Use after spawning subagents; results arrive as next message.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"message\":{\"type\":\"string\"}}}}},{\"type\":\"function\",\"function\":{\"name\":\"subagents\",\"description\":\"List/kill/steer spawned subagents for requester session. If sessions_yield exists, use it for completion; do not poll wait loops.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"enum\":[\"list\",\"kill\",\"steer\"]},\"target\":{\"type\":\"string\"},\"message\":{\"type\":\"string\"},\"recentMinutes\":{\"type\":\"number\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"tts\",\"description\":\"Use only for explicit audio intent (voice/speech/TTS) or active TTS config. Never use for ordinary text replies. Audio auto-delivered from tool result; after success follow reply instructions, no duplicate text/audio.\",\"parameters\":{\"type\":\"object\",\"required\":[\"text\"],\"properties\":{\"text\":{\"type\":\"string\",\"description\":\"Text to speak.\"},\"channel\":{\"type\":\"string\",\"description\":\"Channel id; output-format hint.\"},\"timeoutMs\":{\"type\":\"number\",\"description\":\"Provider timeout ms.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"video_generate\",\"description\":\"Create videos. Session chats: background task; do not call video_generate again for same request; wait completion, then send attachments via message tool. \\\"status\\\" checks active task. Duration may round to provider-supported value.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"action\":{\"type\":\"string\",\"description\":\"\\\"generate\\\" default, \\\"status\\\" active task, \\\"list\\\" providers/models.\"},\"prompt\":{\"type\":\"string\",\"description\":\"Video prompt.\"},\"image\":{\"type\":\"string\",\"description\":\"One reference image path/URL.\"},\"images\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference images; max 9.\"},\"imageRoles\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"`image` + `images` roles by index after de-dupe. Values: first_frame, last_frame, reference_image; empty string leaves unset.\"},\"video\":{\"type\":\"string\",\"description\":\"One reference video path/URL.\"},\"videos\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Reference videos; max 4.\"},\"videoRoles\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"`video` + `videos` roles by index after de-dupe. Value: reference_video; empty string leaves unset.\"},\"model\":{\"type\":\"string\",\"description\":\"Provider/model override, e.g. qwen/wan2.6-t2v.\"},\"filename\":{\"type\":\"string\",\"description\":\"Output filename hint; basename preserved in managed media dir.\"},\"size\":{\"type\":\"string\",\"description\":\"Size hint, e.g. 1280x720, 1920x1080.\"},\"aspectRatio\":{\"type\":\"string\",\"description\":\"Aspect ratio: 1:1, 16:9, 9:16, \\\"adaptive\\\", or provider value; unsupported normalized/ignored.\"},\"resolution\":{\"type\":\"string\",\"description\":\"Resolution: 480P, 720P, 768P, 1080P, 4K, or provider value; unsupported normalized/ignored.\"},\"durationSeconds\":{\"type\":\"number\",\"description\":\"Target seconds; may round to nearest supported duration.\",\"minimum\":1},\"audio\":{\"type\":\"boolean\",\"description\":\"Generated-audio toggle.\"},\"watermark\":{\"type\":\"boolean\",\"description\":\"Watermark toggle.\"},\"providerOptions\":{\"type\":\"object\",\"patternProperties\":{\"^.*$\":{}},\"description\":\"Provider JSON options, e.g. {\\\"seed\\\":42}. Keys/types must match provider capabilities; mismatch skips candidate. Use action=list for accepted keys.\"},\"timeoutMs\":{\"type\":\"number\",\"description\":\"Provider timeout ms.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"web_fetch\",\"description\":\"Fetch URL and extract readable markdown/text. Lightweight page access; no browser automation.\",\"parameters\":{\"type\":\"object\",\"required\":[\"url\"],\"properties\":{\"url\":{\"type\":\"string\",\"description\":\"HTTP(S) URL.\"},\"extractMode\":{\"type\":\"string\",\"enum\":[\"markdown\",\"text\"],\"description\":\"Extract as markdown/text.\",\"default\":\"markdown\"},\"maxChars\":{\"type\":\"number\",\"description\":\"Max chars returned; truncates.\",\"minimum\":100}}}}},{\"type\":\"function\",\"function\":{\"name\":\"web_search\",\"description\":\"Search web for current info; returns normalized provider results.\",\"parameters\":{\"type\":\"object\",\"required\":[\"query\"],\"properties\":{\"query\":{\"type\":\"string\",\"description\":\"Search query.\"},\"count\":{\"type\":\"number\",\"description\":\"Result count.\",\"minimum\":1,\"maximum\":10},\"country\":{\"type\":\"string\",\"description\":\"2-letter country code.\"},\"language\":{\"type\":\"string\",\"description\":\"ISO 639-1 language.\"},\"freshness\":{\"type\":\"string\",\"description\":\"Time filter: day/week/month/year.\"},\"date_after\":{\"type\":\"string\",\"description\":\"Published after YYYY-MM-DD.\"},\"date_before\":{\"type\":\"string\",\"description\":\"Published before YYYY-MM-DD.\"},\"search_lang\":{\"type\":\"string\",\"description\":\"Brave result language.\"},\"ui_lang\":{\"type\":\"string\",\"description\":\"Brave UI locale.\"},\"domain_filter\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"description\":\"Perplexity domain filter.\"},\"max_tokens\":{\"type\":\"number\",\"description\":\"Perplexity total token budget.\",\"minimum\":1,\"maximum\":1000000},\"max_tokens_per_page\":{\"type\":\"number\",\"description\":\"Perplexity tokens per page.\",\"minimum\":1}}}}},{\"type\":\"function\",\"function\":{\"name\":\"write\",\"description\":\"Write content to a file. Creates the file if it doesn't exist, overwrites if it does. Automatically creates parent directories.\",\"parameters\":{\"type\":\"object\",\"required\":[\"path\",\"content\"],\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Path to the file to write (relative or absolute)\"},\"content\":{\"type\":\"string\",\"description\":\"Content to write to the file\"}}}}}],\"tool_choice\":\"auto\"},\"providerResponse\":\"[Streaming - raw response not captured]\",\"response\":{\"content\":\"[Streaming in progress...]\",\"thinking\":null,\"type\":\"streaming\"}}"]], "columns": ["id", "timestamp", "provider", "model", "connectionId", "status", "data"], "primary_keys": ["id"], "primary_key_values": ["1781010923764-l332r6rky"], "units": {}, "query_ms": 1.5134720015339553}