经 AI Skill Hub 精选评估,piclaw Agent工作流 获评「强烈推荐」。这款AI工具在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 8.2 分,适合有一定技术背景的用户使用。
piclaw Agent工作流 是一款基于 TypeScript 开发的开源工具,专注于 AI编码智能体、工作流自动化、开源框架 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。
piclaw Agent工作流 是一款基于 TypeScript 开发的开源工具,专注于 AI编码智能体、工作流自动化、开源框架 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。
# 方式一:npm 全局安装 npm install -g piclaw # 方式二:npx 直接运行(无需安装) npx piclaw --help # 方式三:项目依赖安装 npm install piclaw # 方式四:从源码运行 git clone https://github.com/rcarmo/piclaw cd piclaw npm install npm start
# 命令行使用
piclaw --help
# 基本用法
piclaw [options] <input>
# Node.js 代码中使用
const piclaw = require('piclaw');
const result = await piclaw.run(options);
console.log(result);
# piclaw 配置说明 # 查看配置选项 piclaw --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export PICLAW_CONFIG="/path/to/config.yml"
![]()
Languages: English · 简体中文 · 日本語
PiClaw packages the Pi Coding Agent into a self-hosted workspace with a trilingual streaming web UI, persistent state, multi-provider LLM support, and a practical built-in toolset that includes many add-ons.
It is for people who want one stateful agent workspace they can run locally or in a container without stitching together half a dozen separate services.
bun add -g github:rcarmo/piclaw
Experimental. Linux/macOS/Windows. See docs/install-from-repo.md.
On Windows, PiClaw remains a secondary / not-officially-supported target. Shell-like child processes now run attached there (detached=false) so stdout/stderr remain capturable; Unix-like hosts still use detached process groups for cleaner tree termination on abort/shutdown.
See docs/development.md.
mkdir -p ./home ./workspace
docker run -d \
--init \
--name piclaw \
--restart unless-stopped \
-p 8080:8080 \
-e PICLAW_WEB_PORT=8080 \
-v "$(pwd)/home:/config" \
-v "$(pwd)/workspace:/workspace" \
ghcr.io/rcarmo/piclaw:latest
Open http://localhost:8080 and type /login to configure your LLM provider, including custom OpenAI-compatible endpoints when you are not using one of the built-in hosted providers. The web UI currently ships with English, Simplified Chinese, and Japanese strings; use the Settings language switcher when English is not the hill you want to die on today.
[!TIP] Keep--initenabled fordocker run/podman runso the runtime inserts a tiny init process for signal forwarding and zombie reaping. The bundleddocker-compose.ymlnow sets the equivalentinit: trueflag.
| Mount | Container path | Contents |
|---|---|---|
| Home | /config | Agent home (.pi/, .gitconfig, .bashrc) |
| Workspace | /workspace | Projects, notes, and piclaw state |
[!NOTE] In the container image,/home/agent/.piis backed by/config/.pi. With the stockdocker run/docker-compose.ymlexamples above, Pi home state therefore persists on the host under./home/.pi/agent/. That means provider login state and model metadata should survive rebuilds/recreates when stored under files such as: -./home/.pi/agent/auth.json-./home/.pi/agent/models.jsonMounting directly to/home/agentor/home/agent/.pi/agentcan also work, but/configis the canonical documented persistence path for the container image.
[!WARNING] Never delete /workspace/.piclaw/store/messages.db. It contains chat history, media, and task state.
[!IMPORTANT] You do not need to set provider API keys in piclaw environment variables. PiClaw reuses provider credentials configured in Pi Agent settings.
[!NOTE] Power users can place workspace-scoped shell environment overrides in/workspace/.env.sh. PiClaw sources that file for the embedded terminal and on runtime startup, which is useful for things likePATHtweaks or persistinggh auth loginwithGH_CONFIG_DIR=/workspace/.config/gh. This hook is user-controlled: if its contents break PiClaw startup, shell behavior, or tool resolution, that breakage is the user's responsibility.
Most users only need a few environment variables:
| Variable | Default | Purpose |
|---|---|---|
PICLAW_WEB_PORT | 8080 | Web UI port |
PICLAW_WEB_TERMINAL_ENABLED | 1 on Linux/macOS, 0 on Windows | Enable or disable the authenticated bundled web terminal |
PICLAW_WEB_VNC_ALLOW_DIRECT | 1 on Linux/macOS/Windows | Allow or disable direct VNC targets supplied at runtime |
PICLAW_WEB_TOTP_SECRET | _(empty)_ | Base32 TOTP secret; enables login gate (or initialize with /totp) |
PICLAW_WEB_PASSKEY_MODE | totp-fallback | totp-fallback, passkey-only, or totp-only |
PICLAW_ASSISTANT_NAME | PiClaw | Display name in the UI |
PICLAW_ENABLE_M365_EXPERIMENTAL | 0 | Enable the experimental Microsoft 365 extension bundle |
PICLAW_KEYCHAIN_KEY | _(empty)_ | Master key for encrypted secret storage |
PICLAW_TRUST_PROXY | 0 | Enable when behind a reverse proxy or tunnel |
For the full list, auth setup (TOTP/passkeys), session-scoped SSH-backed remote tools, reverse proxy configuration, SSHFS/FUSE support, and the workspace environment hook, see docs/configuration.md.
综合型AI工作流框架,结合编码智能体与自���应UI交互设计,活跃维护,生态完整,适合专业开发场景。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
AI Skill Hub 点评:piclaw Agent工作流 的核心功能完整,质量优秀。对于AI爱好者来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | piclaw |
| 原始描述 | 开源AI工作流:pi coding agent in a technicolor web trenchcoat 。⭐674 · TypeScript |
| Topics | AI编码智能体工作流自动化开源框架TypeScriptDocker容器 |
| GitHub | https://github.com/rcarmo/piclaw |
| License | MIT |
| 语言 | TypeScript |
收录时间:2026-05-16 · 更新时间:2026-05-19 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。