picoclaw-ai-assistant
reason-machines/trending-skills
Ultra-lightweight AI assistant in Go for $10 hardware with <10MB RAM and multi-provider LLM support.
What is picoclaw-ai-assistant?
PicoClaw is a personal AI assistant written in Go that runs on minimal hardware (Raspberry Pi, LicheeRV-Nano, etc.) with under 10MB RAM and sub-second boot time. It supports multiple LLM providers (OpenAI, Anthropic, Volcengine), optional web search tools, and deploys as a single binary across x86, ARM, MIPS, and RISC-V architectures.
- Deploy a self-contained AI assistant binary on ultra-low-resource devices
- Configure multiple LLM providers (OpenAI-compatible, Anthropic, Volcengine) in a single config
- Enable optional web search via Tavily or Brave Search APIs
- Run interactive chat sessions or send single-message queries from CLI
- Execute agentic tool-call loops with configurable iteration limits
- Support workspace-based file operations and memory management
How to install picoclaw-ai-assistant
npx skills add https://github.com/reason-machines/trending-skills --skill picoclaw-ai-assistant- Go 1.18+ (if building from source)
- API key for at least one LLM provider (OpenAI, Anthropic, or Volcengine)
- Optional: Tavily or Brave Search API key for web search features
- Docker and Docker Compose (if using containerized deployment)
How to use picoclaw-ai-assistant
- 1.Download the precompiled binary for your architecture (ARM64, ARM, RISC-V, x86) from the releases page or build from source with `make build`
- 2.Run `picoclaw onboard` to initialize `~/.picoclaw/config.json` and workspace
- 3.Edit the config file to add your LLM provider API keys (use environment variable references like `$OPENAI_API_KEY`)
- 4.Optionally enable web search tools (Tavily or Brave) by adding API keys and setting `enabled: true`
- 5.Start an interactive session with `picoclaw` or send a single message with `picoclaw -m "your question"`
- 6.Override the default model with `-model <name>` flag if multiple providers are configured
Use cases
- Run a personal coding assistant on a $10 Raspberry Pi Zero 2 W or LicheeRV-Nano
- Deploy a lightweight chatbot on edge devices with minimal power consumption
- Build a multi-model agent that switches between Claude, GPT-4, and Volcengine models
- Set up a web-search-enabled assistant on embedded Linux systems
- Integrate PicoClaw into IoT or robotics projects requiring local AI inference
- Embedded systems and IoT developers
- Hardware enthusiasts building on Raspberry Pi or RISC-V boards
- Developers seeking minimal-footprint AI assistants
- Teams deploying edge AI without cloud dependency
- Makers and hobbyists with resource-constrained devices
picoclaw-ai-assistant FAQ
PicoClaw runs on x86_64, ARM64, ARM (32-bit), MIPS, and RISC-V Linux devices with as little as 10MB RAM. Common targets include Raspberry Pi Zero 2 W, LicheeRV-Nano, and other $10 single-board computers.
No. PicoClaw connects to cloud LLM providers (OpenAI, Anthropic, Volcengine) via API. You provide your API key in the config. Self-hosted models are not mentioned in the documentation.
Yes. Add multiple entries to the `model_list` in config.json and switch between them using the `-model <name>` CLI flag or by setting `model_name` in agent defaults.
Add a Tavily or Brave Search API key to the `tools.web` section in config.json and set `enabled: true`. Only one search provider should be enabled at a time unless you want fallback behavior.
Yes. Use the provided `docker-compose.yml` with profiles for gateway mode (persistent service), launcher mode (web console), or agent mode (one-shot queries).
Full instructions (SKILL.md)
Source of truth, from reason-machines/trending-skills.
name: picoclaw-ai-assistant description: Ultra-lightweight AI assistant in Go that runs on $10 hardware with <10MB RAM, supporting multiple LLM providers, tools, and single-binary deployment across RISC-V, ARM, MIPS, and x86. triggers:
- "set up picoclaw on my device"
- "configure picoclaw with my API key"
- "deploy picoclaw on raspberry pi"
- "build picoclaw from source"
- "add a new LLM provider to picoclaw"
- "run picoclaw in docker"
- "picoclaw web search tool setup"
- "picoclaw memory and workspace configuration"
PicoClaw AI Assistant
Skill by ara.so — Daily 2026 Skills collection.
PicoClaw is an ultra-lightweight personal AI assistant written in Go. It runs on $10 hardware with under 10MB RAM and boots in under 1 second. It supports multiple LLM providers (OpenAI-compatible, Anthropic, Volcengine), optional web search tools, and deploys as a single self-contained binary on x86_64, ARM64, MIPS, and RISC-V Linux devices.
Installation
Precompiled Binary
Download from the releases page:
# Linux ARM64 (Raspberry Pi, LicheeRV-Nano, etc.)
wget https://github.com/sipeed/picoclaw/releases/download/v0.1.1/picoclaw-linux-arm64
chmod +x picoclaw-linux-arm64
./picoclaw-linux-arm64 onboard
Build from Source
git clone https://github.com/sipeed/picoclaw.git
cd picoclaw
# Install dependencies
make deps
# Build for current platform
make build
# Build for all platforms
make build-all
# Raspberry Pi Zero 2 W — 32-bit
make build-linux-arm # → build/picoclaw-linux-arm
# Raspberry Pi Zero 2 W — 64-bit
make build-linux-arm64 # → build/picoclaw-linux-arm64
# Build both Pi Zero variants
make build-pi-zero
# Build and install to system PATH
make install
Docker Compose
git clone https://github.com/sipeed/picoclaw.git
cd picoclaw
# First run — generates docker/data/config.json then exits
docker compose -f docker/docker-compose.yml --profile gateway up
# Edit config
vim docker/data/config.json
# Start in background
docker compose -f docker/docker-compose.yml --profile gateway up -d
# View logs
docker compose -f docker/docker-compose.yml logs -f picoclaw-gateway
# Stop
docker compose -f docker/docker-compose.yml --profile gateway down
Docker: Web Console (Launcher Mode)
docker compose -f docker/docker-compose.yml --profile launcher up -d
# Open http://localhost:18800
Docker: One-shot Agent Mode
# Single question
docker compose -f docker/docker-compose.yml run --rm picoclaw-agent -m "What is 2+2?"
# Interactive session
docker compose -f docker/docker-compose.yml run --rm picoclaw-agent
Docker: Expose Gateway to Host
If the gateway needs to be reachable from the host, set:
PICOCLAW_GATEWAY_HOST=0.0.0.0 docker compose -f docker/docker-compose.yml --profile gateway up -d
Or set PICOCLAW_GATEWAY_HOST=0.0.0.0 in docker/data/config.json.
Termux (Android)
pkg install wget proot
wget https://github.com/sipeed/picoclaw/releases/download/v0.1.1/picoclaw-linux-arm64
chmod +x picoclaw-linux-arm64
termux-chroot ./picoclaw-linux-arm64 onboard
Quick Start
1. Initialize
picoclaw onboard
This creates ~/.picoclaw/config.json with a starter configuration.
2. Configure ~/.picoclaw/config.json
{
"agents": {
"defaults": {
"workspace": "~/.picoclaw/workspace",
"model_name": "gpt-4o",
"max_tokens": 8192,
"temperature": 0.7,
"max_tool_iterations": 20
}
},
"model_list": [
{
"model_name": "gpt-4o",
"model": "openai/gpt-4o",
"api_key": "$OPENAI_API_KEY",
"request_timeout": 300
},
{
"model_name": "claude-sonnet",
"model": "anthropic/claude-sonnet-4-5",
"api_key": "$ANTHROPIC_API_KEY"
},
{
"model_name": "ark-code",
"model": "volcengine/ark-code-latest",
"api_key": "$VOLCENGINE_API_KEY",
"api_base": "https://ark.cn-beijing.volces.com/api/coding/v3"
}
],
"tools": {
"web": {
"brave": {
"enabled": false,
"api_key": "$BRAVE_API_KEY"
},
"tavily": {
"enabled": false,
"api_key": "$TAVILY_API_KEY"
}
}
}
}
Never hard-code API keys. Reference environment variables using
$VAR_NAMEnotation in config, or set them in your shell environment before launch.
3. Run
# Interactive chat
picoclaw
# Single message
picoclaw -m "Summarize the latest Go release notes"
# Use a specific model
picoclaw -model claude-sonnet -m "Refactor this function for clarity"
Key CLI Commands
| Command | Description |
|---|---|
picoclaw onboard | Initialize config and workspace |
picoclaw | Start interactive chat session |
picoclaw -m "..." | Send a single message and exit |
picoclaw -model <name> | Override the default model |
picoclaw -config <path> | Use a custom config file |
Configuration Reference
Model Entry Fields
{
"model_name": "my-model", // Alias used in -model flag and agent defaults
"model": "provider/model-id", // Provider-prefixed model identifier
"api_key": "$ENV_VAR", // API key — use env var reference
"api_base": "https://...", // Optional: override base URL (for self-hosted or regional endpoints)
"request_timeout": 300 // Optional: seconds before timeout
}
Supported Provider Prefixes
| Prefix | Provider |
|---|---|
openai/ | OpenAI and OpenAI-compatible APIs |
anthropic/ | Anthropic Claude |
volcengine/ | Volcengine (Ark) |
Agent Defaults
"agents": {
"defaults": {
"workspace": "~/.picoclaw/workspace", // Working directory for file operations
"model_name": "gpt-4o", // Default model alias
"max_tokens": 8192, // Max response tokens
"temperature": 0.7, // Sampling temperature
"max_tool_iterations": 20 // Max agentic tool-call loop iterations
}
}
Web Search Tools
Get free API keys:
- Tavily: https://tavily.com — 1,000 free queries/month
- Brave Search: https://brave.com/search/api — 2,000 free queries/month
"tools": {
"web": {
"tavily": {
"enabled": true,
"api_key": "$TAVILY_API_KEY"
},
"brave": {
"enabled": false,
"api_key": "$BRAVE_API_KEY"
}
}
}
Only enable one search provider at a time unless you want fallback behavior.
Common Patterns
Pattern: Minimal $10 Device Setup
For a LicheeRV-Nano or similar ultra-low-resource board:
# Download the RISC-V or ARM binary from releases
wget https://github.com/sipeed/picoclaw/releases/download/v0.1.1/picoclaw-linux-riscv64
chmod +x picoclaw-linux-riscv64
# Initialize
./picoclaw-linux-riscv64 onboard
# Edit config — use a lightweight model, low max_tokens
cat > ~/.picoclaw/config.json << 'EOF'
{
"agents": {
"defaults": {
"workspace": "~/.picoclaw/workspace",
"model_name": "gpt-4o-mini",
"max_tokens": 2048,
"temperature": 0.5,
"max_tool_iterations": 10
}
},
"model_list": [
{
"model_name": "gpt-4o-mini",
"model": "openai/gpt-4o-mini",
"api_key": "$OPENAI_API_KEY",
"request_timeout": 120
}
]
}
EOF
./picoclaw-linux-riscv64
Pattern: Full-Stack Dev Assistant with Web Search
{
"agents": {
"defaults": {
"workspace": "~/projects",
"model_name": "claude-sonnet",
"max_tokens": 8192,
"temperature": 0.3,
"max_tool_iterations": 30
}
},
"model_list": [
{
"model_name": "claude-sonnet",
"model": "anthropic/claude-sonnet-4-5",
"api_key": "$ANTHROPIC_API_KEY",
"request_timeout": 600
}
],
"tools": {
"web": {
"tavily": {
"enabled": true,
"api_key": "$TAVILY_API_KEY"
}
}
}
}
Pattern: Docker with Environment Variables
# docker/docker-compose.override.yml
services:
picoclaw-gateway:
environment:
- OPENAI_API_KEY=${OPENAI_API_KEY}
- ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
- TAVILY_API_KEY=${TAVILY_API_KEY}
- PICOCLAW_GATEWAY_HOST=0.0.0.0
export OPENAI_API_KEY=sk-...
export ANTHROPIC_API_KEY=sk-ant-...
docker compose -f docker/docker-compose.yml -f docker/docker-compose.override.yml --profile gateway up -d
Pattern: Build for a Specific Target in Go
# Cross-compile for MIPS (OpenWRT routers)
GOOS=linux GOARCH=mips GOMIPS=softfloat go build -o build/picoclaw-linux-mips ./cmd/picoclaw
# Cross-compile for 32-bit ARM (older Raspberry Pi)
GOOS=linux GOARCH=arm GOARM=7 go build -o build/picoclaw-linux-arm ./cmd/picoclaw
# Cross-compile for RISC-V 64-bit
GOOS=linux GOARCH=riscv64 go build -o build/picoclaw-linux-riscv64 ./cmd/picoclaw
Troubleshooting
Binary won't execute on device
# Verify the binary matches the device architecture
file picoclaw-linux-arm64
uname -m # Should match: aarch64 = arm64, x86_64 = amd64
# Ensure executable permission
chmod +x picoclaw-linux-arm64
"Permission denied" on Termux
Termux requires proot for some system calls:
pkg install proot
termux-chroot ./picoclaw-linux-arm64 onboard
API key not recognized
- Do not use
"api_key": "sk-..."literals in config — set env vars and reference them as"$OPENAI_API_KEY". - Verify the env var is exported in your current shell:
echo $OPENAI_API_KEY.
Docker gateway not reachable from host
Set PICOCLAW_GATEWAY_HOST=0.0.0.0 in the environment or in config.json before starting the container.
High memory usage in recent versions
The project notes that recent PRs may push RAM usage to 10–20MB. If this is a concern on ultra-low-memory devices, pin to an earlier release tag:
git checkout v0.1.1
make build
Config file location
Default: ~/.picoclaw/config.json
Override at runtime:
picoclaw -config /path/to/custom-config.json
Rebuild after dependency changes
cd picoclaw
make deps
make build
Hardware Targets Quick Reference
| Device | Price | Binary |
|---|---|---|
| LicheeRV-Nano (E/W) | ~$10 | picoclaw-linux-riscv64 |
| NanoKVM | ~$30–50 | picoclaw-linux-riscv64 |
| MaixCAM / MaixCAM2 | ~$50–100 | picoclaw-linux-arm64 |
| Raspberry Pi Zero 2 W (32-bit OS) | ~$15 | picoclaw-linux-arm |
| Raspberry Pi Zero 2 W (64-bit OS) | ~$15 | picoclaw-linux-arm64 |
| Android via Termux | — | picoclaw-linux-arm64 |
| Standard Linux x86_64 | — | picoclaw-linux-amd64 |
Related skills
More from reason-machines/trending-skills and the wider catalog.

posterskill-academic-posters
Generate print-ready academic conference posters from Overleaf papers with an interactive drag-and-drop editor.

shannon-ai-pentester
Autonomous white-box AI pentester that analyzes source code and executes live exploits against web apps and APIs.

slug-font-rendering
GPU-accelerated vector font rendering via Slug algorithm with HLSL shaders—no texture atlases needed.

taiwan-md-knowledge-base
AI-native bilingual knowledge base about Taiwan with Markdown SSOT, Astro v5, and D3.js knowledge graph.

tavily-key-generator-proxy
Auto-register Tavily API keys in bulk and pool them behind a unified proxy with round-robin load balancing.

toon-format
Compact, schema-aware JSON encoding that reduces LLM prompt tokens by ~40% while maintaining readability.