masabbs-mcp
MCP server enabling LLMs to retrieve thread-centered discussion context and manage organization design through masabbs REST API.
README
masabbs-mcp
MCP server specification and implementation workspace for connecting LLM clients to masabbs.
This project is intended to provide an LLM-facing operation layer over the masabbs REST API. Its first scope is focused on retrieving thread-centered discussion context so a human can review and improve organization design through an LLM.
Status
v0.2.0 implements a stdio MCP server that connects to masabbs over HTTP REST.
Requirements
- Node.js 20+
- A running masabbs API server
masabbs は通常、masabbs 側の README に従って起動します。
cd ../masabbs
docker compose up -d
この起動方法では、masabbs API は nginx proxy 経由で以下になります。
http://localhost/api/v1
Setup
npm install
Run
MASABBS_BASE_URL=http://localhost/api/v1 npm run dev
After build:
npm run build
MASABBS_BASE_URL=http://localhost/api/v1 node dist/server.js
MCP Client Settings
使用するCLIツールに応じて、MCP設定ファイルに masabbs-mcp を登録します。
Gemini CLI (settings.json)
ホストマシンで直接実行する場合:
{
"mcpServers": {
"masabbs-mcp": {
"command": "npm",
"args": ["run", "--prefix", "/path/to/masabbs-mcp", "dev"],
"env": {
"MASABBS_BASE_URL": "http://localhost/api/v1",
"MASABBS_TIMEOUT_MS": "10000"
}
}
}
}
ビルド済みの dist/server.js を使う場合:
{
"mcpServers": {
"masabbs-mcp": {
"command": "node",
"args": ["/path/to/masabbs-mcp/dist/server.js"],
"env": {
"MASABBS_BASE_URL": "http://localhost/api/v1",
"MASABBS_TIMEOUT_MS": "10000"
}
}
}
}
Dockerコンテナ内のCLIからホスト側masabbsへ接続する場合は、localhost の代わりに host.docker.internal を使います。
{
"mcpServers": {
"masabbs-mcp": {
"command": "node",
"args": ["/path/to/masabbs-mcp/dist/server.js"],
"env": {
"MASABBS_BASE_URL": "http://host.docker.internal/api/v1",
"MASABBS_TIMEOUT_MS": "10000"
}
}
}
}
Codex CLI (config.toml)
ホストマシンで直接実行する場合:
[mcp_servers.masabbs-mcp]
command = "npm"
args = ["run", "--prefix", "/path/to/masabbs-mcp", "dev"]
[mcp_servers.masabbs-mcp.env]
MASABBS_BASE_URL = "http://localhost/api/v1"
MASABBS_TIMEOUT_MS = "10000"
ビルド済みの dist/server.js を使う場合:
[mcp_servers.masabbs-mcp]
command = "node"
args = ["/path/to/masabbs-mcp/dist/server.js"]
[mcp_servers.masabbs-mcp.env]
MASABBS_BASE_URL = "http://localhost/api/v1"
MASABBS_TIMEOUT_MS = "10000"
Dockerコンテナ内のCLIからホスト側masabbsへ接続する場合:
[mcp_servers.masabbs-mcp]
command = "node"
args = ["/path/to/masabbs-mcp/dist/server.js"]
[mcp_servers.masabbs-mcp.env]
MASABBS_BASE_URL = "http://host.docker.internal/api/v1"
MASABBS_TIMEOUT_MS = "10000"
MCP Tools
health_checkget_organizationget_thread_contextget_thread_messagespost_messageget_thread_kpiget_team_kpicreate_teamupdate_teamadd_team_memberremove_team_membercreate_team_relationdelete_team_relationget_team_blueprint
get_thread_context depends on the masabbs endpoint defined in Thread Context REST API Specification:
GET /api/v1/threads/:id/context
Until that masabbs endpoint is implemented, the other tools can still run against existing masabbs REST APIs.
Configuration
| Name | Required | Default | Description |
|---|---|---|---|
MASABBS_BASE_URL |
yes | - | masabbs API base URL, for example http://localhost/api/v1. |
MASABBS_TIMEOUT_MS |
no | 10000 |
HTTP timeout in milliseconds. |
Test
npm run typecheck
npm test
npm run build
Integration tests require a running masabbs API:
MASABBS_BASE_URL=http://localhost:8080/api/v1 npm run test:integration
Integration tests use a lightweight Docker Compose override that exposes the masabbs API server directly on port 8080. This is only for tests. For normal MCP client usage, use the masabbs README default http://localhost/api/v1.
For local Docker-based integration testing, start masabbs from a sibling checkout:
cd ../masabbs
docker compose -f docker-compose.yml -f ../masabbs-mcp/.github/compose.masabbs-it.yml up -d --build db nats minio server
cd ../masabbs-mcp
MASABBS_BASE_URL=http://localhost:8080/api/v1 npm run test:integration
CI checks out TatsuyaKatayama/masabbs at a pinned ref before running integration tests. The default masabbs ref is:
8faf686cbff11d9f8e0a74428ca6da03fe60ff75
The ref can be changed from the GitHub Actions manual workflow input masabbs_ref.
Documents
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.