Tribal is an MCP server implementation for error knowledge tracking and retrieval, providing both REST API and native MCP interfaces for integration with tools like Claude Code and Cline. Developed by Troy Molander at Agentience.ai, it uses ChromaDB for vector similarity search to help AI assistants remember and learn from programming errors, storing error contexts with solutions and finding similar errors via semantic search. The server features JWT authentication, local and cloud storage options, and Docker deployment capabilities, making it ideal for development teams wanting to build a shared knowledge base of programming errors and solutions.
暂无评论. 成为第一个评论的人！
登录以参与讨论
Create new error record. Method: POST /errors
Retrieve error by UUID. Method: GET /errors/{id}
Modify existing error. Method: PUT /errors/{id}
Remove error record. Method: DELETE /errors/{id}
Find errors by criteria. Method: GET /errors
Semantic similarity search for errors. Method: GET /errors/similar
Obtain JWT token for authentication. Method: POST /token
