tribeRef MCP Server

tribeRef MCP Server

Provides persistent tribal memory for Claude Code sessions, enabling querying of decisions and entities, logging artifacts, and managing tasks with Supabase.

Category
访问服务器

README

tribeRef MCP Server

Tribal memory infrastructure for tribeCast AI agents. This is an MCP (Model Context Protocol) server that provides persistent memory across Claude Code sessions.

What tribeRef Does

  1. Tribal Memory (ref_tribal_memory) - Query decisions, entities, schemas, and active tasks
  2. Session Context (ref_session_context) - Track tokens used, items retrieved, session state
  3. Log Artifacts (ref_log_artifact) - Write outputs back to tribal memory
  4. Log Decisions (log_decision) - Record architectural and strategic decisions
  5. Manage Entities (upsert_entity) - Create/update tracked objects (products, services, schemas)
  6. Task Management (create_task, log_completion) - Track work items for agents

Setup

1. Install Dependencies

cd packages/triberef-mcp
npm install

2. Configure Supabase

Copy .env.example to .env and fill in your Supabase credentials:

cp .env.example .env

Required environment variables:

  • SUPABASE_URL - Your Supabase project URL
  • SUPABASE_SERVICE_KEY - Your Supabase service role key

3. Run Database Migration

Execute the migration in your Supabase SQL editor:

  • supabase/migrations/001_tribemcp_foundation.sql

Or use the Supabase CLI.

4. Build

npm run build

5. Run

Stdio mode (for Claude Desktop/CLI):

npm start
# or
node dist/index.cjs

HTTP mode (for browser/API access):

npm run start:http
# or
TRANSPORT=http node dist/index.cjs

Claude Desktop Configuration

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "triberef": {
      "command": "node",
      "args": ["/path/to/tribecast/packages/triberef-mcp/dist/index.cjs"],
      "env": {
        "SUPABASE_URL": "your-supabase-url",
        "SUPABASE_SERVICE_KEY": "your-service-key"
      }
    }
  }
}

Development

# TypeScript watch mode
npm run dev

# Type checking
npm run check

# Build and inspect with MCP inspector
npm run build && npm run inspect

Tools

ref_tribal_memory

Query tribal memory with deduplication (items already retrieved in session are filtered).

{
  "query": "payment architecture",
  "include": ["decisions", "entities", "tasks"],
  "max_tokens": 5000
}

ref_session_context

Get current session state.

{}

Returns: session ID, tokens used/remaining, items retrieved counts.

ref_log_artifact

Log an artifact back to tribal memory.

{
  "task_id": "TRIBE-001",
  "type": "code",
  "name": "src/tools/tribal-memory.ts",
  "content": "..."
}

log_decision

Record an architectural decision.

{
  "topic": "Database architecture",
  "decision": "Use Supabase PostgreSQL",
  "rationale": "Existing infrastructure, real-time subscriptions",
  "tags": ["database", "infrastructure"]
}

upsert_entity

Create or update an entity.

{
  "name": "tribeRef",
  "type": "product",
  "description": "Tribal memory MCP server",
  "status": "building",
  "tags": ["mcp", "infrastructure"]
}

create_task

Create a work item.

{
  "task_id": "TRIBE-003",
  "title": "Implement expert agent consultations",
  "spec": "Add ref_expert_agent tool...",
  "related_entities": ["tribeRef", "tribeAgent"]
}

log_completion

Mark task complete with artifacts.

{
  "task_id": "TRIBE-001",
  "artifacts": [
    { "type": "code", "name": "src/index.ts" },
    { "type": "migration", "name": "001_foundation.sql" }
  ]
}

Architecture

src/
├── index.ts           # MCP server entry point
├── db.ts              # Supabase client & types
├── session.ts         # Session tracking & dedup
├── tools/
│   ├── tribal-memory.ts    # ref_tribal_memory, log_decision, upsert_entity
│   ├── session-context.ts  # ref_session_context
│   └── log-artifact.ts     # ref_log_artifact, create_task, log_completion
└── utils/
    └── tokens.ts      # Token estimation utilities

License

MIT (forked from ref-tools/ref-tools-mcp)

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