PicoClaw: Ultra-Efficient AI Assistant in Go

$10 Hardware · <10MB RAM · <1s Boot · 皮皮虾，我们走！

PicoClaw is an independent open-source project initiated by Sipeed. It is written entirely in Go — not a fork of OpenClaw, NanoBot, or any other project.

🦐 PicoClaw is an ultra-lightweight personal AI Assistant inspired by NanoBot, refactored from the ground up in Go through a self-bootstrapping process, where the AI agent itself drove the entire architectural migration and code optimization.

⚡️ Runs on $10 hardware with <10MB RAM: That's 99% less memory than OpenClaw and 98% cheaper than a Mac mini!

Caution

🚨 SECURITY & OFFICIAL CHANNELS / 安全声明

NO CRYPTO: PicoClaw has NO official token/coin. All claims on pump.fun or other trading platforms are SCAMS.
OFFICIAL DOMAIN: The ONLY official website is picoclaw.io, and company website is sipeed.com
Warning: Many .ai/.org/.com/.net/... domains are registered by third parties.
Warning: picoclaw is in early development now and may have unresolved network security issues. Do not deploy to production environments before the v1.0 release.
Note: picoclaw has recently merged a lot of PRs, which may result in a larger memory footprint (10–20MB) in the latest versions. We plan to prioritize resource optimization as soon as the current feature set reaches a stable state.

📢 News

2026-03-17 🚀 v0.2.3 Released! System tray UI (Windows & Linux), sub-agent status tracking (spawn_status), experimental gateway hot-reload, cron security gates, and 2 security fixes. PicoClaw now at 25K ⭐!

2026-03-09 🎉 v0.2.1 — Biggest update yet! MCP protocol support, 4 new channels (Matrix/IRC/WeCom/Discord Proxy), 3 new providers (Kimi/Minimax/Avian), vision pipeline, JSONL memory store, and model routing.

2026-02-28 📦 v0.2.0 released with Docker Compose support and Web UI launcher.

2026-02-26 🎉 PicoClaw hit 20K stars in just 17 days! Channel auto-orchestration and capability interfaces landed.

Older news...

2026-02-16 🎉 PicoClaw hit 12K stars in one week! Community maintainer roles and roadmap officially posted.

2026-02-13 🎉 PicoClaw hit 5000 stars in 4 days! Project Roadmap and Developer Group setup underway.

2026-02-09 🎉 PicoClaw Launched! Built in 1 day to bring AI Agents to $10 hardware with <10MB RAM. 🦐 PicoClaw，Let's Go！

✨ Features

🪶 Ultra-Lightweight: <10MB Memory footprint — 99% smaller than OpenClaw core functionality.*

💰 Minimal Cost: Efficient enough to run on $10 Hardware — 98% cheaper than a Mac mini.

⚡️ Lightning Fast: 400X Faster startup time, boot in <1 second even on 0.6GHz single core.

🌍 True Portability: Single self-contained binary across RISC-V, ARM, MIPS, and x86, One-click to Go!

🤖 AI-Bootstrapped: Autonomous Go-native implementation — 95% Agent-generated core with human-in-the-loop refinement.

🔌 MCP Support: Native Model Context Protocol integration — connect any MCP server to extend agent capabilities.

👁️ Vision Pipeline: Send images and files directly to the agent — automatic base64 encoding for multimodal LLMs.

🧠 Smart Routing: Rule-based model routing — simple queries go to lightweight models, saving API costs.

*Recent versions may use 10–20MB due to rapid feature merges. Resource optimization is planned. Startup comparison based on 0.8GHz single-core benchmarks (see table below).

	OpenClaw	NanoBot	PicoClaw
Language	TypeScript	Python	Go
RAM	>1GB	>100MB	< 10MB*
Startup (0.8GHz core)	>500s	>30s	<1s
Cost	Mac Mini $599	Most Linux SBC ~$50	Any Linux Board As low as $10

📋 Hardware Compatibility List — See all tested boards, from $5 RISC-V to Raspberry Pi to Android phones. Your board not listed? Submit a PR!

🦾 Demonstration

🛠️ Standard Assistant Workflows

🧩 Full-Stack Engineer	🗂️ Logging & Planning Management	🔎 Web Search & Learning

Develop • Deploy • Scale	Schedule • Automate • Memory	Discovery • Insights • Trends

📱 Run on old Android Phones

Give your decade-old phone a second life! Turn it into a smart AI Assistant with PicoClaw. Quick Start:

Install Termux (Download from GitHub Releases, or search in F-Droid / Google Play).
Execute cmds

# Download the latest release from https://github.com/sipeed/picoclaw/releases
wget https://github.com/sipeed/picoclaw/releases/latest/download/picoclaw_Linux_arm64.tar.gz
tar xzf picoclaw_Linux_arm64.tar.gz
pkg install proot
termux-chroot ./picoclaw onboard   # chroot provides a standard Linux filesystem layout

And then follow the instructions in the "Quick Start" section to complete the configuration!

🐜 Innovative Low-Footprint Deploy

PicoClaw can be deployed on almost any Linux device!

$9.9 LicheeRV-Nano E(Ethernet) or W(WiFi6) version, for Minimal Home Assistant
$30~50 NanoKVM, or $100 NanoKVM-Pro for Automated Server Maintenance
$50 MaixCAM or $100 MaixCAM2 for Smart Monitoring

picoclaw_detect_person.mp4

🌟 More Deployment Cases Await！

📦 Install

Download from picoclaw.io (Recommended)

Visit picoclaw.io — the official website auto-detects your platform and provides one-click download. No need to manually pick an architecture.

Download precompiled binary

Alternatively, download the binary for your platform from the GitHub Releases page.

Build from source (for development)

git clone https://github.com/sipeed/picoclaw.git

cd picoclaw
make deps

# Build, no need to install
make build

# Build for multiple platforms
make build-all

# Build for Raspberry Pi Zero 2 W (32-bit: make build-linux-arm; 64-bit: make build-linux-arm64)
make build-pi-zero

# Build And Install
make install

Raspberry Pi Zero 2 W: Use the binary that matches your OS: 32-bit Raspberry Pi OS → make build-linux-arm; 64-bit → make build-linux-arm64. Or run make build-pi-zero to build both.

📚 Documentation

For detailed guides, see the docs below. The README covers quick start only.

# 1. Clone this repo
git clone https://github.com/sipeed/picoclaw.git
cd picoclaw

# 2. First run — auto-generates docker/data/config.json then exits
docker compose -f docker/docker-compose.yml --profile gateway up
# The container prints "First-run setup complete." and stops.

# 3. Set your API keys
vim docker/data/config.json   # Set provider API keys, bot tokens, etc.

# 4. Start
docker compose -f docker/docker-compose.yml --profile gateway up -d

Tip

Docker Users: By default, the Gateway listens on 127.0.0.1 which is not accessible from the host. If you need to access the health endpoints or expose ports, set PICOCLAW_GATEWAY_HOST=0.0.0.0 in your environment or update config.json.

# 5. Check logs
docker compose -f docker/docker-compose.yml logs -f picoclaw-gateway

# 6. Stop
docker compose -f docker/docker-compose.yml --profile gateway down

Launcher Mode (Web Console)

The launcher image includes all three binaries (picoclaw, picoclaw-launcher, picoclaw-launcher-tui) and starts the web console by default, which provides a browser-based UI for configuration and chat.

docker compose -f docker/docker-compose.yml --profile launcher up -d

Open http://localhost:18800 in your browser. The launcher manages the gateway process automatically.

Warning

The web console does not yet support authentication. Avoid exposing it to the public internet.

Agent Mode (One-shot)

# Ask a question
docker compose -f docker/docker-compose.yml run --rm picoclaw-agent -m "What is 2+2?"

# Interactive mode
docker compose -f docker/docker-compose.yml run --rm picoclaw-agent

Update

docker compose -f docker/docker-compose.yml pull
docker compose -f docker/docker-compose.yml --profile gateway up -d

🚀 Quick Start

Tip

Set your API Key in ~/.picoclaw/config.json. Get API Keys: Volcengine (CodingPlan) (LLM) · OpenRouter (LLM) · Zhipu (LLM). Web search is optional — get a free Tavily API (1000 free queries/month) or Brave Search API (2000 free queries/month).

1. Initialize

picoclaw onboard

2. Configure (~/.picoclaw/config.json)

{
  "agents": {
    "defaults": {
      "workspace": "~/.picoclaw/workspace",
      "model_name": "gpt-5.4",
      "max_tokens": 8192,
      "temperature": 0.7,
      "max_tool_iterations": 20
    }
  },
  "model_list": [
    {
      "model_name": "ark-code-latest",
      "model": "volcengine/ark-code-latest",
      "api_key": "sk-your-api-key"
    },
    {
      "model_name": "gpt-5.4",
      "model": "openai/gpt-5.4",
      "api_key": "your-api-key",
      "request_timeout": 300
    },
    {
      "model_name": "claude-sonnet-4.6",
      "model": "anthropic/claude-sonnet-4.6",
      "api_key": "your-anthropic-key"
    }
  ],
  "tools": {
    "web": {
      "brave": {
        "enabled": false,
        "api_key": "YOUR_BRAVE_API_KEY",
        "max_results": 5
      },
      "tavily": {
        "enabled": false,
        "api_key": "YOUR_TAVILY_API_KEY",
        "max_results": 5
      },
      "duckduckgo": {
        "enabled": true,
        "max_results": 5
      },
      "perplexity": {
        "enabled": false,
        "api_key": "YOUR_PERPLEXITY_API_KEY",
        "max_results": 5
      },
      "searxng": {
        "enabled": false,
        "base_url": "http://your-searxng-instance:8888",
        "max_results": 5
      }
    }
  }
}

New: The model_list configuration format allows zero-code provider addition. See Model Configuration for details. request_timeout is optional and uses seconds. If omitted or set to <= 0, PicoClaw uses the default timeout (120s).

3. Get API Keys

LLM Provider: OpenRouter · Zhipu · Anthropic · OpenAI · Gemini
Web Search (optional):
- Brave Search - Paid ($5/1000 queries, ~$5-6/month)
- Perplexity - AI-powered search with chat interface
- SearXNG - Self-hosted metasearch engine (free, no API key needed)
- Tavily - Optimized for AI Agents (1000 requests/month)
- DuckDuckGo - Built-in fallback (no API key required)

Note: See config.example.json for a complete configuration template.

4. Chat

picoclaw agent -m "What is 2+2?"

That's it! You have a working AI assistant in 2 minutes.

💬 Chat Apps

Talk to your picoclaw through Telegram, Discord, WhatsApp, Matrix, QQ, DingTalk, LINE, or WeCom

Note: All webhook-based channels (LINE, WeCom, etc.) are served on a single shared Gateway HTTP server (gateway.host:gateway.port, default 127.0.0.1:18790). There are no per-channel ports to configure. Note: Feishu uses WebSocket/SDK mode and does not use the shared HTTP webhook server.

Channel	Setup
Telegram	Easy (just a token)
Discord	Easy (bot token + intents)
WhatsApp	Easy (native: QR scan; or bridge URL)
Weixin	Easy (Native QR scan)
Matrix	Medium (homeserver + bot access token)
QQ	Easy (AppID + AppSecret)
DingTalk	Medium (app credentials)
LINE	Medium (credentials + webhook URL)
WeCom AI Bot	Medium (Token + AES key)

Telegram (Recommended)

1. Create a bot

Open Telegram, search @BotFather
Send /newbot, follow prompts
Copy the token

2. Configure

{
  "channels": {
    "telegram": {
      "enabled": true,
      "token": "YOUR_BOT_TOKEN",
      "allow_from": ["YOUR_USER_ID"]
    }
  }
}

Get your user ID from @userinfobot on Telegram.

3. Run

picoclaw gateway

4. Telegram command menu (auto-registered at startup)

PicoClaw now keeps command definitions in one shared registry. On startup, Telegram will automatically register supported bot commands (for example /start, /help, /show, /list) so command menu and runtime behavior stay in sync. Telegram command menu registration remains channel-local discovery UX; generic command execution is handled centrally in the agent loop via the commands executor.

If command registration fails (network/API transient errors), the channel still starts and PicoClaw retries registration in the background.

Discord

1. Create a bot

Go to https://discord.com/developers/applications
Create an application → Bot → Add Bot
Copy the bot token

2. Enable intents

In the Bot settings, enable MESSAGE CONTENT INTENT
(Optional) Enable SERVER MEMBERS INTENT if you plan to use allow lists based on member data

3. Get your User ID

Discord Settings → Advanced → enable Developer Mode
Right-click your avatar → Copy User ID

4. Configure

{
  "channels": {
    "discord": {
      "enabled": true,
      "token": "YOUR_BOT_TOKEN",
      "allow_from": ["YOUR_USER_ID"]
    }
  }
}

5. Invite the bot

OAuth2 → URL Generator
Scopes: bot
Bot Permissions: Send Messages, Read Message History
Open the generated invite URL and add the bot to your server

Optional: Group trigger mode

By default the bot responds to all messages in a server channel. To restrict responses to @-mentions only, add:

{
  "channels": {
    "discord": {
      "group_trigger": { "mention_only": true }
    }
  }
}

You can also trigger by keyword prefixes (e.g. !bot):

{
  "channels": {
    "discord": {
      "group_trigger": { "prefixes": ["!bot"] }
    }
  }
}

6. Run

picoclaw gateway

WhatsApp (native via whatsmeow)

PicoClaw can connect to WhatsApp in two ways:

Native (recommended): In-process using whatsmeow. No separate bridge. Set "use_native": true and leave bridge_url empty. On first run, scan the QR code with WhatsApp (Linked Devices). Session is stored under your workspace (e.g. workspace/whatsapp/). The native channel is optional to keep the default binary small; build with -tags whatsapp_native (e.g. make build-whatsapp-native or go build -tags whatsapp_native ./cmd/...).
Bridge: Connect to an external WebSocket bridge. Set bridge_url (e.g. ws://localhost:3001) and keep use_native false.

Configure (native)

{
  "channels": {
    "whatsapp": {
      "enabled": true,
      "use_native": true,
      "session_store_path": "",
      "allow_from": []
    }
  }
}

If session_store_path is empty, the session is stored in <workspace>/whatsapp/. Run picoclaw gateway; on first run, scan the QR code printed in the terminal with WhatsApp → Linked Devices.

Weixin (WeChat Personal)

PicoClaw supports connecting to your personal WeChat account using the official Tencent iLink API.

1. Login Run the interactive QR login flow:

picoclaw onboard weixin

Scan the printed QR code with your WeChat mobile app. On success, the token is saved to your config.

2. Configure (Optional) Update allow_from with your WeChat User ID to restrict who can message the bot:

{
  "channels": {
    "weixin": {
      "enabled": true,
      "token": "YOUR_TOKEN",
      "allow_from": ["YOUR_USER_ID"]
    }
  }
}

3. Run

picoclaw gateway

QQ

1. Create a bot

Go to QQ Open Platform
Create an application → Get AppID and AppSecret

2. Configure

{
  "channels": {
    "qq": {
      "enabled": true,
      "app_id": "YOUR_APP_ID",
      "app_secret": "YOUR_APP_SECRET",
      "allow_from": []
    }
  }
}

Set allow_from to empty to allow all users, or specify QQ numbers to restrict access.

3. Run

picoclaw gateway

DingTalk

1. Create a bot

Go to Open Platform
Create an internal app
Copy Client ID and Client Secret

2. Configure

{
  "channels": {
    "dingtalk": {
      "enabled": true,
      "client_id": "YOUR_CLIENT_ID",
      "client_secret": "YOUR_CLIENT_SECRET",
      "allow_from": []
    }
  }
}

Set allow_from to empty to allow all users, or specify DingTalk user IDs to restrict access.

3. Run

picoclaw gateway

Matrix

1. Prepare bot account

Use your preferred homeserver (e.g. https://matrix.org or self-hosted)
Create a bot user and obtain its access token

2. Configure

{
  "channels": {
    "matrix": {
      "enabled": true,
      "homeserver": "https://matrix.org",
      "user_id": "@your-bot:matrix.org",
      "access_token": "YOUR_MATRIX_ACCESS_TOKEN",
      "allow_from": []
    }
  }
}

3. Run

picoclaw gateway

For full options (device_id, join_on_invite, group_trigger, placeholder, reasoning_channel_id), see Matrix Channel Configuration Guide.

LINE

1. Create a LINE Official Account

Go to LINE Developers Console
Create a provider → Create a Messaging API channel
Copy Channel Secret and Channel Access Token

2. Configure

{
  "channels": {
    "line": {
      "enabled": true,
      "channel_secret": "YOUR_CHANNEL_SECRET",
      "channel_access_token": "YOUR_CHANNEL_ACCESS_TOKEN",
      "webhook_path": "/webhook/line",
      "allow_from": []
    }
  }
}

LINE webhook is served on the shared Gateway server (gateway.host:gateway.port, default 127.0.0.1:18790).

3. Set up Webhook URL

LINE requires HTTPS for webhooks. Use a reverse proxy or tunnel:

# Example with ngrok (gateway default port is 18790)
ngrok http 18790

Then set the Webhook URL in LINE Developers Console to https://your-domain/webhook/line and enable Use webhook.

4. Run

picoclaw gateway

In group chats, the bot responds only when @mentioned. Replies quote the original message.

WeCom (企业微信)

PicoClaw supports three types of WeCom integration:

Option 1: WeCom Bot (Bot) - Easier setup, supports group chats Option 2: WeCom App (Custom App) - More features, proactive messaging, private chat only Option 3: WeCom AI Bot (AI Bot) - Official AI Bot, streaming replies, supports group & private chat

See WeCom AI Bot Configuration Guide for detailed setup instructions.

Quick Setup - WeCom AI Bot:

1. Create an AI Bot

Go to WeCom Admin Console → AI Bot
Create a new AI Bot → Set name, avatar, etc.
Copy Bot ID and Secret

2. Configure

{
  "channels": {
    "wecom_aibot": {
      "enabled": true,
      "bot_id": "YOUR_BOT_ID",
      "secret": "YOUR_SECRET",
      "allow_from": [],
      "welcome_message": "Hello! How can I help you?"
    }
  }
}

3. Run

picoclaw gateway

Note: WeCom AI Bot uses streaming pull protocol — no reply timeout concerns. Long tasks (>30 seconds) automatically switch to response_url push delivery.

Join the Agent Social Network

Connect Picoclaw to the Agent Social Network simply by sending a single message via the CLI or any integrated Chat App.

Read https://clawdchat.ai/skill.md and follow the instructions to join ClawdChat.ai

⚙️ Configuration

Config file: ~/.picoclaw/config.json

Environment Variables

You can override default paths using environment variables. This is useful for portable installations, containerized deployments, or running picoclaw as a system service. These variables are independent and control different paths.

Variable	Description	Default Path
`PICOCLAW_CONFIG`	Overrides the path to the configuration file. This directly tells picoclaw which `config.json` to load, ignoring all other locations.	`~/.picoclaw/config.json`
`PICOCLAW_HOME`	Overrides the root directory for picoclaw data. This changes the default location of the `workspace` and other data directories.	`~/.picoclaw`

Examples:

# Run picoclaw using a specific config file
# The workspace path will be read from within that config file
PICOCLAW_CONFIG=/etc/picoclaw/production.json picoclaw gateway

# Run picoclaw with all its data stored in /opt/picoclaw
# Config will be loaded from the default ~/.picoclaw/config.json
# Workspace will be created at /opt/picoclaw/workspace
PICOCLAW_HOME=/opt/picoclaw picoclaw agent

# Use both for a fully customized setup
PICOCLAW_HOME=/srv/picoclaw PICOCLAW_CONFIG=/srv/picoclaw/main.json picoclaw gateway

Workspace Layout

PicoClaw stores data in your configured workspace (default: ~/.picoclaw/workspace):

~/.picoclaw/workspace/
├── sessions/          # Conversation sessions and history
├── memory/            # Long-term memory (MEMORY.md)
├── state/             # Persistent state (last channel, etc.)
├── cron/              # Scheduled jobs database
├── skills/            # Workspace-specific skills
├── AGENT.md           # Structured agent definition and system prompt
├── SOUL.md            # Agent soul
├── USER.md            # User profile and preferences for this workspace
├── HEARTBEAT.md       # Periodic task prompts (checked every 30 min)
└── ...

Skill Sources

By default, skills are loaded from:

~/.picoclaw/workspace/skills (workspace)
~/.picoclaw/skills (global)
<current-working-directory>/skills (builtin)

For advanced/test setups, you can override the builtin skills root with:

export PICOCLAW_BUILTIN_SKILLS=/path/to/skills

Unified Command Execution Policy

Generic slash commands are executed through a single path in pkg/agent/loop.go via commands.Executor.
Channel adapters no longer consume generic commands locally; they forward inbound text to the bus/agent path. Telegram still auto-registers supported commands at startup.
Unknown slash command (for example /foo) passes through to normal LLM processing.
Registered but unsupported command on the current channel (for example /show on WhatsApp) returns an explicit user-facing error and stops further processing.

🔒 Security Sandbox

PicoClaw runs in a sandboxed environment by default. The agent can only access files and execute commands within the configured workspace.

Default Configuration

{
  "agents": {
    "defaults": {
      "workspace": "~/.picoclaw/workspace",
      "restrict_to_workspace": true
    }
  }
}

Option	Default	Description
`workspace`	`~/.picoclaw/workspace`	Working directory for the agent
`restrict_to_workspace`	`true`	Restrict file/command access to workspace

Protected Tools

When restrict_to_workspace: true, the following tools are sandboxed:

Tool	Function	Restriction
`read_file`	Read files	Only files within workspace
`write_file`	Write files	Only files within workspace
`list_dir`	List directories	Only directories within workspace
`edit_file`	Edit files	Only files within workspace
`append_file`	Append to files	Only files within workspace
`exec`	Execute commands	Command paths must be within workspace

Additional Exec Protection

Even with restrict_to_workspace: false, the exec tool blocks these dangerous commands:

rm -rf, del /f, rmdir /s — Bulk deletion
format, mkfs, diskpart — Disk formatting
dd if= — Disk imaging
Writing to /dev/sd[a-z] — Direct disk writes
shutdown, reboot, poweroff — System shutdown
Fork bomb :(){ :|:& };:

Error Examples

[ERROR] tool: Tool execution failed
{tool=exec, error=Command blocked by safety guard (path outside working dir)}

[ERROR] tool: Tool execution failed
{tool=exec, error=Command blocked by safety guard (dangerous pattern detected)}

Disabling Restrictions (Security Risk)

If you need the agent to access paths outside the workspace:

Method 1: Config file

{
  "agents": {
    "defaults": {
      "restrict_to_workspace": false
    }
  }
}

Method 2: Environment variable

export PICOCLAW_AGENTS_DEFAULTS_RESTRICT_TO_WORKSPACE=false

⚠️ Warning: Disabling this restriction allows the agent to access any path on your system. Use with caution in controlled environments only.

Security Boundary Consistency

The restrict_to_workspace setting applies consistently across all execution paths:

Execution Path	Security Boundary
Main Agent	`restrict_to_workspace` ✅
Subagent / Spawn	Inherits same restriction ✅
Heartbeat tasks	Inherits same restriction ✅

All paths share the same workspace restriction — there's no way to bypass the security boundary through subagents or scheduled tasks.

Heartbeat (Periodic Tasks)

PicoClaw can perform periodic tasks automatically. Create a HEARTBEAT.md file in your workspace:

# Periodic Tasks

- Check my email for important messages
- Review my calendar for upcoming events
- Check the weather forecast

The agent will read this file every 30 minutes (configurable) and execute any tasks using available tools.

Async Tasks with Spawn

For long-running tasks (web search, API calls), use the spawn tool to create a subagent:

# Periodic Tasks

## Quick Tasks (respond directly)

- Report current time

## Long Tasks (use spawn for async)

- Search the web for AI news and summarize
- Check email and report important messages

Key behaviors:

Feature	Description
spawn	Creates async subagent, doesn't block heartbeat
Independent context	Subagent has its own context, no session history
message tool	Subagent communicates with user directly via message tool
Non-blocking	After spawning, heartbeat continues to next task

How Subagent Communication Works

Heartbeat triggers
    ↓
Agent reads HEARTBEAT.md
    ↓
For long task: spawn subagent
    ↓                           ↓
Continue to next task      Subagent works independently
    ↓                           ↓
All tasks done            Subagent uses "message" tool
    ↓                           ↓
Respond HEARTBEAT_OK      User receives result directly

The subagent has access to tools (message, web_search, etc.) and can communicate with the user independently without going through the main agent.

Configuration:

{
  "heartbeat": {
    "enabled": true,
    "interval": 30
  }
}

Option	Default	Description
`enabled`	`true`	Enable/disable heartbeat
`interval`	`30`	Check interval in minutes (min: 5)

Environment variables:

PICOCLAW_HEARTBEAT_ENABLED=false to disable
PICOCLAW_HEARTBEAT_INTERVAL=60 to change interval

Providers

Note

Groq provides free voice transcription via Whisper. If configured, audio messages from any channel will be automatically transcribed at the agent level.

Provider	Purpose	Get API Key
`gemini`	LLM (Gemini direct)	aistudio.google.com
`zhipu`	LLM (Zhipu direct)	bigmodel.cn
`volcengine`	LLM(Volcengine direct)	volcengine.com
`openrouter`	LLM (recommended, access to all models)	openrouter.ai
`anthropic`	LLM (Claude direct)	console.anthropic.com
`openai`	LLM (GPT direct)	platform.openai.com
`deepseek`	LLM (DeepSeek direct)	platform.deepseek.com
`qwen`	LLM (Qwen direct)	dashscope.console.aliyun.com
`groq`	LLM + Voice transcription (Whisper)	console.groq.com
`cerebras`	LLM (Cerebras direct)	cerebras.ai
`vivgrid`	LLM (Vivgrid direct)	vivgrid.com

Model Configuration (model_list)

What's New? PicoClaw now uses a model-centric configuration approach. Simply specify vendor/model format (e.g., zhipu/glm-4.7) to add new providers—zero code changes required!

This design also enables multi-agent support with flexible provider selection:

Different agents, different providers: Each agent can use its own LLM provider
Model fallbacks: Configure primary and fallback models for resilience
Load balancing: Distribute requests across multiple endpoints
Centralized configuration: Manage all providers in one place

📋 All Supported Vendors

Vendor	`model` Prefix	Default API Base	Protocol	API Key
OpenAI	`openai/`	`https://api.openai.com/v1`	OpenAI	Get Key
Anthropic	`anthropic/`	`https://api.anthropic.com/v1`	Anthropic	Get Key
智谱 AI (GLM)	`zhipu/`	`https://open.bigmodel.cn/api/paas/v4`	OpenAI	Get Key
DeepSeek	`deepseek/`	`https://api.deepseek.com/v1`	OpenAI	Get Key
Google Gemini	`gemini/`	`https://generativelanguage.googleapis.com/v1beta`	OpenAI	Get Key
Groq	`groq/`	`https://api.groq.com/openai/v1`	OpenAI	Get Key
Moonshot	`moonshot/`	`https://api.moonshot.cn/v1`	OpenAI	Get Key
通义千问 (Qwen)	`qwen/`	`https://dashscope.aliyuncs.com/compatible-mode/v1`	OpenAI	Get Key
NVIDIA	`nvidia/`	`https://integrate.api.nvidia.com/v1`	OpenAI	Get Key
Ollama	`ollama/`	`http://localhost:11434/v1`	OpenAI	Local (no key needed)
OpenRouter	`openrouter/`	`https://openrouter.ai/api/v1`	OpenAI	Get Key
LiteLLM Proxy	`litellm/`	`http://localhost:4000/v1`	OpenAI	Your LiteLLM proxy key
VLLM	`vllm/`	`http://localhost:8000/v1`	OpenAI	Local
Cerebras	`cerebras/`	`https://api.cerebras.ai/v1`	OpenAI	Get Key
VolcEngine (Doubao)	`volcengine/`	`https://ark.cn-beijing.volces.com/api/v3`	OpenAI	Get Key
神算云	`shengsuanyun/`	`https://router.shengsuanyun.com/api/v1`	OpenAI	-
BytePlus	`byteplus/`	`https://ark.ap-southeast.bytepluses.com/api/v3`	OpenAI	Get Key
Vivgrid	`vivgrid/`	`https://api.vivgrid.com/v1`	OpenAI	Get Key
LongCat	`longcat/`	`https://api.longcat.chat/openai`	OpenAI	Get Key
ModelScope (魔搭)	`modelscope/`	`https://api-inference.modelscope.cn/v1`	OpenAI	Get Token
Antigravity	`antigravity/`	Google Cloud	Custom	OAuth only
GitHub Copilot	`github-copilot/`	`localhost:4321`	gRPC	-

Basic Configuration

{
  "model_list": [
    {
      "model_name": "ark-code-latest",
      "model": "volcengine/ark-code-latest",
      "api_key": "sk-your-api-key"
    },
    {
      "model_name": "gpt-5.4",
      "model": "openai/gpt-5.4",
      "api_key": "sk-your-openai-key"
    },
    {
      "model_name": "claude-sonnet-4.6",
      "model": "anthropic/claude-sonnet-4.6",
      "api_key": "sk-ant-your-key"
    },
    {
      "model_name": "glm-4.7",
      "model": "zhipu/glm-4.7",
      "api_key": "your-zhipu-key"
    }
  ],
  "agents": {
    "defaults": {
      "model": "gpt-5.4"
    }
  }
}

Vendor-Specific Examples

OpenAI

{
  "model_name": "gpt-5.4",
  "model": "openai/gpt-5.4",
  "api_key": "sk-..."
}

VolcEngine (Doubao)

{
  "model_name": "ark-code-latest",
  "model": "volcengine/ark-code-latest",
  "api_key": "sk-..."
}

智谱 AI (GLM)

{
  "model_name": "glm-4.7",
  "model": "zhipu/glm-4.7",
  "api_key": "your-key"
}

DeepSeek

{
  "model_name": "deepseek-chat",
  "model": "deepseek/deepseek-chat",
  "api_key": "sk-..."
}

Anthropic (with API key)

{
  "model_name": "claude-sonnet-4.6",
  "model": "anthropic/claude-sonnet-4.6",
  "api_key": "sk-ant-your-key"
}

Run picoclaw auth login --provider anthropic to paste your API token.

Anthropic Messages API (native format)

For direct Anthropic API access or custom endpoints that only support Anthropic's native message format:

{
  "model_name": "claude-opus-4-6",
  "model": "anthropic-messages/claude-opus-4-6",
  "api_key": "sk-ant-your-key",
  "api_base": "https://api.anthropic.com"
}

Use anthropic-messages protocol when:

Using third-party proxies that only support Anthropic's native /v1/messages endpoint (not OpenAI-compatible /v1/chat/completions)

Connecting to services like MiniMax, Synthetic that require Anthropic's native message format

The existing anthropic protocol returns 404 errors (indicating the endpoint doesn't support OpenAI-compatible format)

Note: The anthropic protocol uses OpenAI-compatible format (/v1/chat/completions), while anthropic-messages uses Anthropic's native format (/v1/messages). Choose based on your endpoint's supported format.

Ollama (local)

{
  "model_name": "llama3",
  "model": "ollama/llama3"
}

Custom Proxy/API

{
  "model_name": "my-custom-model",
  "model": "openai/custom-model",
  "api_base": "https://my-proxy.com/v1",
  "api_key": "sk-...",
  "request_timeout": 300
}

LiteLLM Proxy

{
  "model_name": "lite-gpt4",
  "model": "litellm/lite-gpt4",
  "api_base": "http://localhost:4000/v1",
  "api_key": "sk-..."
}

PicoClaw strips only the outer litellm/ prefix before sending the request, so proxy aliases like litellm/lite-gpt4 send lite-gpt4, while litellm/openai/gpt-4o sends openai/gpt-4o.

Load Balancing

Configure multiple endpoints for the same model name—PicoClaw will automatically round-robin between them:

{
  "model_list": [
    {
      "model_name": "gpt-5.4",
      "model": "openai/gpt-5.4",
      "api_base": "https://api1.example.com/v1",
      "api_key": "sk-key1"
    },
    {
      "model_name": "gpt-5.4",
      "model": "openai/gpt-5.4",
      "api_base": "https://api2.example.com/v1",
      "api_key": "sk-key2"
    }
  ]
}

Migration from Legacy `providers` Config

The old providers configuration is deprecated but still supported for backward compatibility.

Old Config (deprecated):

{
  "providers": {
    "zhipu": {
      "api_key": "your-key",
      "api_base": "https://open.bigmodel.cn/api/paas/v4"
    }
  },
  "agents": {
    "defaults": {
      "provider": "zhipu",
      "model": "glm-4.7"
    }
  }
}

New Config (recommended):

{
  "model_list": [
    {
      "model_name": "glm-4.7",
      "model": "zhipu/glm-4.7",
      "api_key": "your-key"
    }
  ],
  "agents": {
    "defaults": {
      "model": "glm-4.7"
    }
  }
}

For detailed migration guide, see docs/migration/model-list-migration.md.

Provider Architecture

PicoClaw routes providers by protocol family:

OpenAI-compatible protocol: OpenRouter, OpenAI-compatible gateways, Groq, Zhipu, and vLLM-style endpoints.
Anthropic protocol: Claude-native API behavior.
Codex/OAuth path: OpenAI OAuth/token authentication route.

This keeps the runtime lightweight while making new OpenAI-compatible backends mostly a config operation (api_base + api_key).

Zhipu

1. Get API key and base URL

Get API key

2. Configure

{
  "agents": {
    "defaults": {
      "workspace": "~/.picoclaw/workspace",
      "model": "glm-4.7",
      "max_tokens": 8192,
      "temperature": 0.7,
      "max_tool_iterations": 20
    }
  },
  "providers": {
    "zhipu": {
      "api_key": "Your API Key",
      "api_base": "https://open.bigmodel.cn/api/paas/v4"
    }
  }
}

3. Run

picoclaw agent -m "Hello"

Full config example

{
  "agents": {
    "defaults": {
      "model": "anthropic/claude-opus-4-5"
    }
  },
  "session": {
    "dm_scope": "per-channel-peer",
    "backlog_limit": 20
  },
  "providers": {
    "openrouter": {
      "api_key": "sk-or-v1-xxx"
    },
    "groq": {
      "api_key": "gsk_xxx"
    }
  },
  "channels": {
    "telegram": {
      "enabled": true,
      "token": "123456:ABC...",
      "allow_from": ["123456789"]
    },
    "discord": {
      "enabled": true,
      "token": "",
      "allow_from": [""]
    },
    "whatsapp": {
      "enabled": false,
      "bridge_url": "ws://localhost:3001",
      "use_native": false,
      "session_store_path": "",
      "allow_from": []
    },
    "feishu": {
      "enabled": false,
      "app_id": "cli_xxx",
      "app_secret": "xxx",
      "encrypt_key": "",
      "verification_token": "",
      "allow_from": []
    },
    "qq": {
      "enabled": false,
      "app_id": "",
      "app_secret": "",
      "allow_from": []
    }
  },
  "tools": {
    "web": {
      "brave": {
        "enabled": false,
        "api_key": "BSA...",
        "max_results": 5
      },
      "duckduckgo": {
        "enabled": true,
        "max_results": 5
      },
      "perplexity": {
        "enabled": false,
        "api_key": "",
        "max_results": 5
      },
      "searxng": {
        "enabled": false,
        "base_url": "http://localhost:8888",
        "max_results": 5
      }
    },
    "cron": {
      "exec_timeout_minutes": 5
    }
  },
  "heartbeat": {
    "enabled": true,
    "interval": 30
  }
}

🖥️ CLI Reference

Command	Description
`picoclaw onboard`	Initialize config & workspace
`picoclaw onboard weixin`	Connect WeChat account via QR
`picoclaw agent -m "..."`	Chat with the agent
`picoclaw agent`	Interactive chat mode
`picoclaw gateway`	Start the gateway
`picoclaw status`	Show status
`picoclaw version`	Show version info
`picoclaw model`	Show or change default model
`picoclaw cron list`	List all scheduled jobs
`picoclaw cron add ...`	Add a scheduled job
`picoclaw cron disable`	Disable a scheduled job
`picoclaw cron remove`	Remove a scheduled job
`picoclaw skills list`	List installed skills
`picoclaw skills install`	Install a skill
`picoclaw migrate`	Migrate data from older versions
`picoclaw auth login`	Authenticate with providers

Scheduled Tasks / Reminders

PicoClaw supports scheduled reminders and recurring tasks through the cron tool:

One-time reminders: "Remind me in 10 minutes" → triggers once after 10min
Recurring tasks: "Remind me every 2 hours" → triggers every 2 hours
Cron expressions: "Remind me at 9am daily" → uses cron expression

🤝 Contribute & Roadmap

PRs welcome! The codebase is intentionally small and readable. 🤗

See our full Community Roadmap.

Developer group building, join after your first merged PR!

User Groups:

discord: https://discord.gg/V4sAZ9XWpN

WeChat:

Name		Name	Last commit message	Last commit date
Latest commit History 1,362 Commits
.github		.github
assets		assets
cmd		cmd
config		config
docker		docker
docs		docs
examples/pico-echo-server		examples/pico-echo-server
pkg		pkg
scripts		scripts
web		web
workspace		workspace
.dockerignore		.dockerignore
.env.example		.env.example
.gitignore		.gitignore
.golangci.yaml		.golangci.yaml
.goreleaser.yaml		.goreleaser.yaml
CONTRIBUTING.md		CONTRIBUTING.md
CONTRIBUTING.zh.md		CONTRIBUTING.zh.md
Dockerfile.custom		Dockerfile.custom
LICENSE		LICENSE
Makefile		Makefile
README.fr.md		README.fr.md
README.id.md		README.id.md
README.it.md		README.it.md
README.ja.md		README.ja.md
README.md		README.md
README.pt-br.md		README.pt-br.md
README.vi.md		README.vi.md
README.zh.md		README.zh.md
ROADMAP.md		ROADMAP.md
go.mod		go.mod
go.sum		go.sum
render.yaml		render.yaml

Folders and files

Latest commit

History

Repository files navigation

PicoClaw: Ultra-Efficient AI Assistant in Go

$10 Hardware · <10MB RAM · <1s Boot · 皮皮虾，我们走！

📢 News

✨ Features

🦾 Demonstration

🛠️ Standard Assistant Workflows

📱 Run on old Android Phones

🐜 Innovative Low-Footprint Deploy

📦 Install

Download from picoclaw.io (Recommended)

Download precompiled binary

Build from source (for development)

📚 Documentation

Launcher Mode (Web Console)

Agent Mode (One-shot)

Update

🚀 Quick Start

💬 Chat Apps

Join the Agent Social Network

⚙️ Configuration

Environment Variables

Workspace Layout

Skill Sources

Unified Command Execution Policy

🔒 Security Sandbox

Default Configuration

Protected Tools

Additional Exec Protection

Error Examples

Disabling Restrictions (Security Risk)

Security Boundary Consistency

Heartbeat (Periodic Tasks)

Async Tasks with Spawn

How Subagent Communication Works

Providers

Model Configuration (model_list)

📋 All Supported Vendors

Basic Configuration

Vendor-Specific Examples

Load Balancing

Migration from Legacy providers Config

Provider Architecture

🖥️ CLI Reference

Scheduled Tasks / Reminders

🤝 Contribute & Roadmap

About

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Migration from Legacy `providers` Config

Packages