picoclaw-ai-assistant
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.
What this skill does
# PicoClaw AI Assistant
> Skill by [ara.so](https://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](https://github.com/sipeed/picoclaw/releases):
```bash
# 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
```bash
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
```bash
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)
```bash
docker compose -f docker/docker-compose.yml --profile launcher up -d
# Open http://localhost:18800
```
#### Docker: One-shot Agent Mode
```bash
# 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:
```bash
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)
```bash
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
```bash
picoclaw onboard
```
This creates `~/.picoclaw/config.json` with a starter configuration.
### 2. Configure `~/.picoclaw/config.json`
```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 `$VAR_NAME` notation in config, or set them in your shell environment before launch.
### 3. Run
```bash
# Interactive chat
picoclaw
# Single message
picoclaw -m "Summarize the latest Go release notes"
# Use a specific model
picoclaw -model claude-sonnet -m "Refactor this function for clarity"
```
---
## Key CLI Commands
| Command | Description |
|---|---|
| `picoclaw onboard` | Initialize config and workspace |
| `picoclaw` | Start interactive chat session |
| `picoclaw -m "..."` | Send a single message and exit |
| `picoclaw -model <name>` | Override the default model |
| `picoclaw -config <path>` | Use a custom config file |
---
## Configuration Reference
### Model Entry Fields
```json
{
"model_name": "my-model", // Alias used in -model flag and agent defaults
"model": "provider/model-id", // Provider-prefixed model identifier
"api_key": "$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
```json
"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
```json
"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 $10 Device Setup
For a LicheeRV-Nano or similar ultra-low-resource board:
```bash
# Download the RISC-V or ARM binary from releases
wget https://github.com/sipeed/picoclaw/releases/download/v0.1.1/picoclaw-linux-riscv64
chmod +x picoclaw-linux-riscv64
# Initialize
./picoclaw-linux-riscv64 onboard
# Edit config — use a lightweight model, low max_tokens
cat > ~/.picoclaw/config.json << 'EOF'
{
"agents": {
"defaults": {
"workspace": "~/.picoclaw/workspace",
"model_name": "gpt-4o-mini",
"max_tokens": 2048,
"temperature": 0.5,
"max_tool_iterations": 10
}
},
"model_list": [
{
"model_name": "gpt-4o-mini",
"model": "openai/gpt-4o-mini",
"api_key": "$OPENAI_API_KEY",
"request_timeout": 120
}
]
}
EOF
./picoclaw-linux-riscv64
```
### Pattern: Full-Stack Dev Assistant with Web Search
```json
{
"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
```yaml
# docker/docker-compose.override.yml
services:
picoclaw-gateway:
environment:
- OPENAI_API_KEY=${OPENAI_API_KEY}
- ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
- TAVILY_API_KEY=${TAVILY_API_KEY}
- PICOCLAW_GATEWAY_HOST=0.0.0.0
```
```bash
export OPENAI_API_KEY=sk-...
export ANTHROPIC_API_KEY=sk-ant-...
docker compose -f docker/docker-compose.yml -f docker/docker-compose.override.yml --profile gateway up -d
```
### Pattern: Build Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.