# Pi — Minimal Terminal Coding Harness You Own

> Pi is a minimal terminal coding harness you customize with extensions, skills, and prompts, so you can run a consistent personal or team agent shell.

## Install

Copy the content below into your project:

## Quick Use

1. Install:
   ```bash
   npm install -g @earendil-works/pi-coding-agent
   # or
   curl -fsSL https://pi.dev/install.sh | sh
   ```
2. Start in a repo:
   ```bash
   cd /path/to/repo
   pi
   ```
3. Authenticate (choose one):
   ```bash
   # subscription login inside pi
   /login
   
   # or API key
   export ANTHROPIC_API_KEY=sk-ant-...
   pi
   ```

## Intro

Pi is a minimal terminal coding harness you customize with extensions, skills, and prompts, so you can run a consistent personal or team agent shell.

- **Best for:** developers who want an agent harness they can script and extend (skills, prompts, UI) across many projects
- **Works with:** Node.js + npm; interactive mode plus one-shot `-p` prompts; supports context files like `AGENTS.md`
- **Setup time:** 15–30 minutes

## Practical Notes

- Use `AGENTS.md` per repo to encode rules (tests, migrations, output format).
- Start with one workflow (lint/test/fix) and add extensions only when needed.

## Why a “harness” matters

Most coding agents ship as a fixed product. Pi is intentionally small so you can:
- encode team rules in context files,
- add skills/prompts as reusable building blocks,
- keep the workflow consistent across repos.

## Recommended setup for teams

### 1) Add project instructions

Create `AGENTS.md` in the repo root:

```md
- Run `npm test` after code changes.
- Do not modify production migrations.
- Keep PR summaries under 10 lines.
```

### 2) Teach Pi your repo’s “definition of done”

Ask Pi to:
1) find the relevant scripts (`package.json`, `Makefile`, etc.),
2) run checks,
3) propose the smallest diff,
4) re-run checks.

## One-shot mode for automation

Pi supports non-interactive prompts:

```bash
pi -p "Summarize this repo and tell me how to run checks."
```

This is useful for CI assistants or scheduled hygiene tasks.

### FAQ

**Q: Is Pi a model?**
A: No. It’s a harness that connects to providers and exposes tools, skills, and customization.

**Q: How do I keep behavior consistent across repos?**
A: Use shared skills/prompts and repo-local context files (`AGENTS.md`).

**Q: Can I use it non-interactively?**
A: Yes. Use `pi -p "..."` for one-shot prompts in scripts or CI.

## Source & Thanks

> Source: https://github.com/earendil-works/pi
> License: MIT
> GitHub stars: 48,265 · forks: 5,740

---

<!-- ZH -->

## 快速使用

1. 安装：
   ```bash
   npm install -g @earendil-works/pi-coding-agent
   # 或
   curl -fsSL https://pi.dev/install.sh | sh
   ```
2. 在仓库中启动：
   ```bash
   cd /path/to/repo
   pi
   ```
3. 认证（二选一）：
   ```bash
   # 在 pi 内用订阅登录
   /login
   
   # 或使用 API key
   export ANTHROPIC_API_KEY=sk-ant-...
   pi
   ```

## 简介

Pi 是一个极简的终端编码 harness：通过扩展、skills 与 prompt 模板自定义成你的个人/团队 agent 外壳，并用上下文文件固化规则，避免被某个固定 UX 绑死与被动迁移。

- **适合谁：** 想把 agent 工作流脚本化、并能跨项目复用扩展（skills/prompt/UI）的开发者
- **可搭配：** Node.js + npm；支持交互模式与一次性 `-p`；支持 `AGENTS.md` 等上下文文件
- **准备时间：** 15–30 分钟

## 实战建议

- 在每个仓库用 `AGENTS.md` 写清规则（测试、迁移、输出格式等）。
- 先跑通一个核心工作流（lint/test/fix），再按需增加扩展。

## 为什么“harness”很重要

很多 coding agent 是固定产品形态；Pi 刻意保持核心很小，便于你：
- 用上下文文件固化团队规则，  
- 把 skills/prompts 做成可复用积木，  
- 跨仓库保持一致的交付标准。

## 团队推荐落地方式

### 1）写清项目规则

在仓库根目录加 `AGENTS.md`：

```md
- 改代码后必须跑 `npm test`。
- 不要修改生产迁移脚本。
- PR 总结不超过 10 行。
```

### 2）把“完成标准”教给 Pi

让 Pi 按固定节奏做事：
1）找项目脚本（`package.json`/`Makefile` 等），  
2）运行检查，  
3）给出最小 diff，  
4）再跑一次检查验证。

## 一次性模式用于自动化

Pi 也支持非交互 prompt：

```bash
pi -p "Summarize this repo and tell me how to run checks."
```

适合 CI 助手或定时的仓库 hygiene 任务。

### FAQ

**Pi 是模型吗？**
答：不是。它是一个连接多种 provider 的 harness，并暴露 tools/skills/自定义能力。

**怎么跨仓库保持一致行为？**
答：复用共享 skills/prompts，并在每个仓库用 `AGENTS.md` 固化规则。

**能非交互使用吗？**
答：可以。用 `pi -p "..."` 做一次性调用，适合脚本或 CI。

## 来源与感谢

> Source: https://github.com/earendil-works/pi
> License: MIT
> GitHub stars: 48,265 · forks: 5,740


---
Source: https://tokrepo.com/en/workflows/pi-minimal-terminal-coding-harness-you-own
Author: Agent Toolkit