briefcase

briefcase

Enables AI tools to retrieve codebase context, summaries, decision notes, and file details via MCP, avoiding re-explaining the project.

Category
访问服务器

README

Briefcase

Carry your codebase's context wherever you code.

Never re explain your project to an AI again. Briefcase scans your codebase, generates compact AI summaries, tracks changes via git, stores your "why" decisions, and exports a paste ready briefing for Claude, Cursor, Codex, or any AI tool.

Why Briefcase?

Every time you switch AI tools or accounts, you lose context. Briefcase maintains a living briefing document about your project tech stack, file summaries, architecture decisions so the next AI understands your codebase in seconds.

Install

cd briefcase
pip install -e .

Set your Anthropic API key for AI powered summaries (optional heuristic fallback works without it):

export ANTHROPIC_API_KEY=sk-ant-...

Quick Start

Inside any project folder:

briefcase init      # Set up .briefcase/ in this project
briefcase scan      # Scan and summarize all significant files
briefcase export    # Copy-paste context into your AI chat

Commands

Command Description
briefcase init Initialize Briefcase in the current project
briefcase scan Full scan summarize all significant files
briefcase update Incremental update only changed files (via git)
briefcase export Output paste-ready briefing markdown
briefcase note "..." Save a decision note (use -f path to attach to a file)
briefcase notes List all decision notes
briefcase log Show scan history timeline
briefcase diff Show briefing changes since last export
briefcase status Health check — how fresh is your briefing?
briefcase focus <path> Deep-dive export for one file/folder
briefcase serve Run as MCP server for direct AI tool integration

How It Works

briefcase scan / update
        │
        ▼
   scanner.py ──► walks project, detects tech stack, skips junk
        │
        ▼
  git_tracker.py ──► finds changed files since last scan
        │
        ▼
  summarizer.py ──► Claude API (or heuristics) per file
        │
        ▼
   storage.py ──► SQLite in .briefcase/briefing.db
        │
        ▼
   exporter.py ──► paste-ready markdown context block

What gets stored

your-project/
└── .briefcase/
    ├── briefing.db    # SQLite — summaries, notes, history
    └── config.json    # Tech stack, settings

MCP Server (Stretch)

Run Briefcase as an MCP server so Cursor or Claude can pull context directly:

briefcase serve

Add to your Cursor MCP config:

{
  "mcpServers": {
    "briefcase": {
      "command": "briefcase",
      "args": ["serve"],
      "cwd": "/path/to/your/project"
    }
  }
}

Available MCP tools: get_briefing, get_briefing_diff, get_project_overview, get_notes, get_file_summary.

Examples

# First time in a project
briefcase init
briefcase scan

# After coding for a while
briefcase update
briefcase export -o CONTEXT.md

# Remember why you made a choice
briefcase note "Chose SQLite over Postgres — single-user, no server needed" -f briefcase/storage.py
briefcase notes

# Before switching AI tools
briefcase diff
briefcase status

Development

pip install -e ".[dev]"
pytest

Project Structure

briefcase/
├── briefcase/
│   ├── cli.py           # CLI commands
│   ├── scanner.py       # Project walker + tech stack detection
│   ├── summarizer.py    # Claude API / heuristic summaries
│   ├── git_tracker.py   # Git diff integration
│   ├── storage.py       # SQLite persistence
│   ├── notes.py         # Decision notes
│   ├── exporter.py      # Paste-ready export
│   └── mcp_server.py    # MCP server mode
├── tests/
├── pyproject.toml
└── README.md

License

MIT

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选