# Project Docs

> 维护有边界的 BLUEPRINT、ARCHITECTURE 和 CURRENT_HANDOFF 文档，并运行 Docs Lint。适用于重要实施或诊断、得到授权的交接、范围或架构变化，以及文档漂移检查；普通小修不触发。

- **Type:** Skill
- **Install:** `agentstack add skill-candycong2001-ui-agent-project-playbook-project-docs`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [candycong2001-ui](https://agentstack.voostack.com/s/candycong2001-ui)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [candycong2001-ui](https://github.com/candycong2001-ui)
- **Source:** https://github.com/candycong2001-ui/agent-project-playbook/tree/main/templates/zh-CN/skills/project-docs

## Install

```sh
agentstack add skill-candycong2001-ui-agent-project-playbook-project-docs
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# 项目文档

## 流程

1. 开始重要实施或诊断时，先读 `docs/CURRENT_HANDOFF.md`，再按需读稳定文档的相关部分。
2. 用当前仓库证据核对文档内容。
3. 探索或只读审查期间，未经用户授权只报告漂移，不修改文档。
4. 任务开始时，替换 `CURRENT_HANDOFF.md` 的当前目标和下一步。
5. 得到可验证成果后，替换当前状态和验证证据，不追加时间线。
6. 交接前删除已经完成的过程，只留下可以继续工作的单一当前状态。
7. 只有产品意图、用户、范围、约束或验收变化时才更新 `BLUEPRINT.md`。
8. 只有组件、边界、数据流、已验证命令、部署、回滚或风险变化时才更新 `ARCHITECTURE.md`。
9. 重要交付前运行 `python3 .agents/skills/project-docs/scripts/docs_lint.py`。

## 边界

- 当前代码、测试、配置、命令输出和用户指令高于这些文档。
- BLUEPRINT 不超过 160 行。
- ARCHITECTURE 不超过 160 行。
- CURRENT_HANDOFF 不超过 80 行。
- AGENTS 不超过 90 行。
- Git 保存历史；CURRENT_HANDOFF 只保存最新工作状态。

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [candycong2001-ui](https://github.com/candycong2001-ui)
- **Source:** [candycong2001-ui/agent-project-playbook](https://github.com/candycong2001-ui/agent-project-playbook)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-candycong2001-ui-agent-project-playbook-project-docs
- Seller: https://agentstack.voostack.com/s/candycong2001-ui
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.