componecat

componecat

Enables AI agents to access an organization's software component catalog, including entities, relationships, interfaces, documentation, and team ownership, for grounded context and impact analysis.

Category
访问服务器

README

Componecat MCP Server

Componecat is a software component catalog: a hierarchical, richly-typed registry of every system, service, and library your organization operates — with structured metadata, Git integration, hosted documentation, endpoint definitions, dependency relationships, and team ownership. Its MCP server gives AI agents the organizational context they're usually missing, so they can answer with your architecture instead of guessing.

This repository is the public home of the server's MCP registry metadata (server.json) and the @componecat/mcp stdio bridge. The server itself is part of the hosted Componecat app:

https://app.componecat.ai/api/mcp

Registry name: ai.componecat/componecat · Transport: Streamable HTTP · Website: componecat.ai · Docs: docs.componecat.ai

Catalogs are private and organization-scoped, so the endpoint requires authentication (see Authorization). You'll need a Componecat account — sign up at app.componecat.ai.

Install

Install in VS Code Install in VS Code Insiders

VS Code

Click an install badge above, or add to your mcp.json:

{
  "servers": {
    "componecat": {
      "type": "http",
      "url": "https://app.componecat.ai/api/mcp"
    }
  }
}

Claude Code

claude mcp add --transport http componecat https://app.componecat.ai/api/mcp

Claude.ai / Claude Desktop

Add a custom connector: Settings → Connectors → Add custom connector, with URL https://app.componecat.ai/api/mcp.

Cursor

Add to Cursor, or add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "componecat": {
      "url": "https://app.componecat.ai/api/mcp"
    }
  }
}

ChatGPT

Add a connector in Settings → Connectors (developer mode) with the server URL https://app.componecat.ai/api/mcp.

Stdio-only clients

For clients that only speak stdio, the bridge in this repo proxies to the remote endpoint via mcp-remote (OAuth completes in your browser):

{
  "mcpServers": {
    "componecat": {
      "command": "npx",
      "args": ["-y", "@componecat/mcp"]
    }
  }
}

Self-hosted Componecat instance? Point the bridge at it with COMPONECAT_MCP_URL=https://your-instance.example.com/api/mcp.

Authorization

Componecat implements the standard MCP authorization flow with OAuth 2.1:

  • OAuth (authorization code + PKCE) — the usual path for interactive clients. The client discovers the authorization server automatically; dynamic client registration is supported, so most clients need no manual setup. Device flow is available for clients that can't open a browser.
  • Personal access tokens — bearer tokens for scripts and non-interactive clients.

Access is scoped: you grant an agent read or write access per resource category (catalog entities, teams, schema), and can scope a token down to specific entities. A read-only grant is enough for an agent that only needs context.

Tools

60 tools, each annotated with read-only / destructive / idempotent hints so clients can gate confirmation appropriately.

Search & browse global_search · search_catalog (full-text with typo tolerance, or a structured query language — field:value, AND/OR/NOT, comparisons, ranges) · list_catalog_entities · get_catalog_entity · list_by_owner · list_entity_kind_definitions · get_current_organization

Relationships & impact analysis get_relationships · get_relationship_kinds · add_relationship · remove_relationship · get_impact_analysis (transitive downstream impact of a change)

Interfaces & endpoints get_interfaces · create_interface · update_interface · delete_interface · import_interface_spec (import OpenAPI/AsyncAPI specs)

Documentation get_documentation · create_documentation_section · update_documentation_section · delete_documentation_section · list_documentation_links · create_documentation_link · update_documentation_link · delete_documentation_link

Git-sourced documentation (sections synced from Markdown in your repos) configure_git_sourced_documentation · list_git_sourced_documentation · update_git_sourced_documentation · remove_git_sourced_documentation · sync_git_sourced_documentation

Activity log list_activity_log_entries · add_activity_log_entry · delete_activity_log_entry

Catalog maintenance create_catalog_entity · update_catalog_entity · delete_catalog_entity · update_entity_field_values

Schema definitions (entity kinds, custom fields, relationship kinds) create_entity_kind_definition · update_entity_kind_definition · delete_entity_kind_definition · list_entity_field_definitions · create_entity_field_definition · update_entity_field_definition · delete_entity_field_definition · add_kind_field_assignment · update_kind_field_assignment · remove_kind_field_assignment · list_relationship_kind_definitions · create_relationship_kind_definition · update_relationship_kind_definition · delete_relationship_kind_definition

Teams & ownership list_teams · get_team · create_team · update_team · delete_team · add_team_members · remove_team_member

Product documentation list_componecat_documentation · read_componecat_documentation — Componecat's own docs, so an agent can walk you through setup instead of guessing.

The server also exposes the catalog as MCP resources: every entity is readable as its componecat.yaml descriptor, and a kind's entities can be listed without knowing the taxonomy in advance.

Typical workflows

  • Grounded code generation — before generating code for a service, the agent looks up its kind, languages, interfaces, and conventions.
  • Cross-service integration — the agent follows depends-on edges, reads the target's interface spec, and writes a correct client.
  • Impact analysis — before a change, the agent traverses transitive relationships to see what breaks downstream.
  • Ownership resolution — the agent finds the owning team to route a review or an escalation.
  • Documentation-grounded answers — runbooks and architecture notes answer questions with your context, not generic guesswork.

Links

License

The contents of this repository (registry metadata and the stdio bridge) are MIT licensed. The Componecat application itself is a hosted product and is not open source.

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