MCP Debug Recorder

MCP Debug Recorder

Record debug sessions, commands, failed attempts, and successful fixes in local SQLite storage.

Category
访问服务器

README

mcp-debug-recorder

npm version License npm downloads LobeHub Glama

mcp-debug-recorder answers a simple question fast: have I fixed this before?

It records debug sessions, terminal commands, failed attempts, and successful fixes in a local SQLite database so your MCP client can query your own debugging history in natural language.

Quick Start

npx mcp-debug-recorder

By default, data is stored at ~/.mcp-debug-recorder/sessions.db.

Architecture

src/
├── db.ts           - openDb(), createTestDb(), versioned MIGRATIONS[]
├── store.ts        - Store class with dependency-injected SQLite access
├── search.ts       - FTS5 + Fuse.js hybrid search
├── tools/          - MCP tool handlers grouped by session/search/admin concerns
├── types.ts        - Zod schemas and TypeScript types
├── mcp.ts          - MCP server wiring + tool registration
├── server-http.ts  - Streamable HTTP transport
├── logging.ts      - Structured logging with secret redaction
└── version.ts      - Package version helper

Schema versioning

The database schema is versioned via PRAGMA user_version. Migrations run automatically on startup, so upgrading does not require manual SQL.

Adding a custom database path

DEBUG_RECORDER_DB=/path/to/custom.db npx mcp-debug-recorder

Configuration

Environment variables

  • DEBUG_RECORDER_DB: override the SQLite database path
  • PORT: override the HTTP server port for Streamable HTTP mode
  • LOG_LEVEL: minimum structured log level (debug, info, warn, error)
  • FUZZY_THRESHOLD: override the Fuse.js threshold used during reranking

Available Tools

  • start_debug_session: start tracking a new issue
  • add_fix: record a failed or successful fix attempt
  • record_command: save a terminal command and its output
  • close_session: mark a session as resolved or abandoned
  • update_session: edit title, description, or tags
  • delete_session: permanently delete a session with explicit confirmation
  • search_sessions: search historical sessions with FTS5 + fuzzy reranking
  • find_similar_errors: ask whether you have seen a similar error before
  • get_session: fetch full session details
  • get_session_context: fetch an AI-friendly summary of a session
  • list_sessions: browse sessions with filters
  • get_stats: summarize your debug history
  • export_sessions: export your local history for backup or migration
  • import_sessions: import a previously exported JSON payload

Client Setup

Claude Desktop

{
  "mcpServers": {
    "mcp-debug-recorder": {
      "command": "npx",
      "args": ["mcp-debug-recorder"]
    }
  }
}

VS Code / GitHub Copilot

Create or update .vscode/mcp.json:

{
  "servers": {
    "mcp-debug-recorder": {
      "type": "stdio",
      "command": "npx",
      "args": ["mcp-debug-recorder"]
    }
  }
}

Codex CLI

codex mcp add mcp-debug-recorder -- npx mcp-debug-recorder
codex mcp list

Gemini CLI

gemini mcp add mcp-debug-recorder npx mcp-debug-recorder
gemini mcp list

Antigravity

antigravity --add-mcp "{\"name\":\"mcp-debug-recorder\",\"command\":\"npx\",\"args\":[\"mcp-debug-recorder\"]}"

Real Usage Examples

Have I seen this before?

"I'm getting TypeError: Cannot read properties of undefined, have I seen this before?"

Call find_similar_errors with the current error text, then inspect the best match with get_session_context.

Record an active incident

  1. Call start_debug_session
  2. Add terminal commands with record_command
  3. Add each attempted fix with add_fix
  4. Use update_session when the title or notes become clearer
  5. Close the session with close_session

Back up your local debug history

  1. Call export_sessions with format: "json"
  2. Save the returned JSON in your preferred backup system
  3. Restore later with import_sessions

Data Storage

  • Default path: ~/.mcp-debug-recorder/sessions.db
  • Portable SQLite storage with better-sqlite3
  • FTS5-backed search index for large histories
  • No external database server required

Note: better-sqlite3 uses a native addon. If you see binding errors, run npm rebuild better-sqlite3 for your Node version.

HTTP Transport

The package also supports Streamable HTTP:

npm run start:http

Useful routes:

  • GET /health
  • GET /version
  • MCP endpoint: POST/GET/DELETE /mcp

Development

npm ci
npm run lint
npm test
npm run build
npm run test:e2e
npm run docs:api

For release verification:

npm run format:check
npm run test:coverage
npm run prepublishOnly

Additional project docs:

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