PicoClaw

PicoClaw is an open-source command-line coding agent developed by Sipeed and community contributors. Written in Go, it is engineered for lightweight footprint, fast execution, and direct terminal-driven software development workflows. In day-to-day operation, PicoClaw interacts with codebases directly through the CLI, orchestrating automated file inspection, editing, and task resolution. It supports custom OpenAI-compatible API endpoints, specialized providers such as NEAR AI Cloud, and popular LLM backends. Developers can configure fallback model chains directly to ensure continuity when primary providers experience rate limits or outages. PicoClaw differentiates itself through its minimal resource overhead and streamlined configuration management. Instead of relying on heavy graphical frontends or complex multi-container setups, it offers native CLI commands like `picoclaw model add` to discover and register remote or local LLM endpoints dynamically from base URLs. Distributed as open source under the Sipeed GitHub repository, PicoClaw is free to use and modify under standard open-source licensing, requiring only user-provided API keys or local inference endpoints.

Tags: AI, ai agent, cli, developer tools, OpenSource, productivity

Visit PicoClaw

Key features

  • OpenAI-compatible Endpoint Support — Connect to any OpenAI-compatible API or local LLM server using standard base URL and key flags. - Dynamic Model Discovery — Query remote endpoints via CLI commands like picoclaw model add to inspect available models and set defaults. - Configurable Fallback Chains — Set up multi-provider model fallback sequences to gracefully handle rate limits or API downtime. - Multi-Provider Ecosystem — Native support for cloud AI providers including NEAR AI Cloud alongside standard LLM backends. - Go-Based Architecture — Implemented in Go for fast startup times, low system memory usage, and simple single-binary distribution. - Docker Deployment Options — Includes containerized setup files for isolated environment execution and repeatable development pipelines.

Use cases

Use Case: Connecting Local and Self-Hosted LLMs Developers running private inference servers with vLLM or Ollama can register their custom OpenAI-compatible endpoints using the CLI. PicoClaw queries the server's model list, binds the configuration, and executes coding tasks against the private infrastructure.

Use Case: Automated Fallback Orchestration Engineers working with production API quotas can configure fallback model chains. If primary cloud models hit rate limits or downtime during code generation, PicoClaw automatically cascades requests to secondary providers without interrupting the terminal session.

Use Case: Lightweight Terminal-Based Refactoring Developers working over SSH on remote or resource-constrained machines can run PicoClaw as a single binary to inspect repository files, generate edits, and run commands without graphical or heavyweight IDE dependencies.

Who it is for

PicoClaw is designed for developers, systems engineers, and power users who prefer minimal, fast CLI tools and want direct control over custom or self-hosted LLM endpoints. It is less suitable for users looking for integrated visual IDE extensions or graphical chat interfaces.

Install guide

Installation Build from source using Go (standard Go installation method) or clone the repository: bash git clone https://github.com/sipeed/picoclaw.git cd picoclaw go build -o picoclaw ./cmd/picoclaw

First Run Add your preferred OpenAI-compatible model endpoint and API key to register your default model: bash # Add custom OpenAI-compatible provider ./picoclaw model add -b https://api.openai.com/v1 -k YOUR_API_KEY -m gpt-4o -n default # Or add a local endpoint ./picoclaw model add -b http://localhost:8000/v1 -k none -m local-model -n local

Example Session Run PicoClaw directly from your project directory to execute tasks: bash # Inspect codebase and implement a feature picoclaw "Add unit tests for the HTTP response parser in cmd/"

Part of CLI Coding Agents

Categories: AI Coding Agents, Assistant, Developer Tools, Productivity

Related CLI agents

  • GreatCTO — GreatCTO is an open-source AI agent orchestration harness and engineering pipeline framework created by Anatoly Velikiy (avelikiy). It coordinates multi-role…
  • Agent Island — Agent Island is an open-source agent orchestration harness created by developer tristan666666 on GitHub. Designed as a framework for running…
  • Playwright Praman — Playwright Praman is an open-source test harness and agent orchestrator developed by Maheshwar Kanitkar. It is designed to bridge terminal…
  • Stratless — Stratless is an open-source agent harness and execution orchestrator developed by the Stratless AI team. It provides a modular infrastructure…
  • Skillreaper — Skillreaper is an open-source skill lifecycle management and auditing tool for AI agent ecosystems, developed by thousandflowers. It analyzes agent…
  • Zosma QA — Zosma QA is an open-source testing-focused agent harness and orchestration framework created by Zosma AI. It provides scaffolding, structured prompt…
  • AI Tools
  • Categories
  • Industries
  • CLI Coding Agents
  • MCP Servers
  • MCP Categories