slack-mcp
MCP server that exposes Slack Web API operations as tools, enabling users to list conversations, read history, post messages, and call any Slack Web API method.
README
slack-mcp
Slack MCP server built from the skeleton-mcp base.
This project exposes Slack Web API operations as MCP tools with:
- stdio and HTTP transport support
- HTTP bearer-token protection for /mcp
- optional admin authorization gates for mutating tools
- Vault/Postgres scaffolding preserved from the original skeleton
What It Does
The server currently focuses on Slack operations:
- Validate Slack token context
- List conversations
- Read conversation history
- Read user profile info
- Post messages
- Call any Slack Web API method via a generic tool
Slack MCP Tool Catalog
Read-only tools:
- slack_connection_info
- slack_scope_info
- slack_list_methods
- slack_auth_test
- slack_conversations_list
- slack_conversation_history
- slack_user_info
Mutating tools:
- slack_post_message
- slack_api_call (only guarded when the method is mutating)
Authorization behavior:
- If MCP_ADMIN_AUTH_KEY is set, mutating tools require authorizationKey.
- Guarded mutating method names for slack_api_call include chat.postMessage and common write operations like reactions.add, pins.add, conversations.archive, etc.
Slack API Notes
Configured Slack client:
- Base URL defaults to https://slack.com/api
- Bot token is required via SLACK_BOT_TOKEN
Implemented method wrappers:
- auth.test
- conversations.list
- conversations.history
- users.info
- chat.postMessage
Generic method execution:
- slack_api_call sends POST to /<methodName> with params JSON payload
- Example methodName: conversations.info, users.list, chat.postMessage
Required Slack Scopes
Typical scopes you should grant your Slack app for these tools:
- chat:write for slack_post_message
- channels:read, groups:read, im:read, mpim:read for conversations.list
- channels:history, groups:history, im:history, mpim:history for conversations.history
- users:read for users.info
- users:read.email if you need email in users.info responses
Exact required scopes can vary by workspace and conversation type.
Environment Variables
Slack-specific:
- SLACK_BOT_TOKEN (required)
- SLACK_API_BASE_URL (default: https://slack.com/api)
- SLACK_TIMEOUT_MS (default: 15000)
Core app:
- APP_NAME
- MCP_SERVER_NAME (default: slack-mcp)
- MCP_SERVER_VERSION
- MCP_ADMIN_AUTH_KEY
- MCP_TRANSPORT_MODE (stdio, http, both)
HTTP transport:
- MCP_HTTP_HOST
- MCP_HTTP_PORT
- MCP_HTTP_PATH
- MCP_HTTP_HEALTH_PATH
- MCP_HTTP_AUTH_MODE
- MCP_HTTP_AUTH_TOKENS
- MCP_HTTP_TRUST_PROXY
- MCP_HTTP_ALLOWED_ORIGINS
- MCP_HTTP_ALLOWED_IPS
- MCP_HTTP_MAX_BODY_BYTES
- MCP_HTTP_RATE_LIMIT_WINDOW_MS
- MCP_HTTP_RATE_LIMIT_MAX_REQUESTS
For full defaults and examples, see .env.example.
Quick Start
- Install dependencies:
npm install
- Copy env file and set Slack token:
cp .env.example .env
Set at minimum:
- SLACK_BOT_TOKEN=xoxb-...
- Start MCP server:
npm run start:stdio
Or HTTP mode:
npm run start:http
- Run tests:
npm test
Registering in MCP Clients
VS Code (stdio)
Example .vscode/mcp.json entry:
{
"command": "npm",
"args": ["run", "start:stdio"],
"cwd": "/path/to/slack-mcp"
}
HTTP-capable clients
Run:
npm run start:http
Default endpoint:
- MCP: http://127.0.0.1:3000/mcp
- Health: http://127.0.0.1:3000/healthz
HTTP Security Model
- Every /mcp request requires Authorization: Bearer <token>
- Token validation currently uses configured MCP_HTTP_AUTH_TOKENS unless other auth sources are wired
- Rate limit and body-size protections are enforced by src/http/server.js
Note:
- TLS termination is intended at reverse proxy/load balancer level in this process mode.
External Services Mode
The repository still supports an app-only startup profile via docker-compose.external.yml when using external Vault and Postgres services.
Required variables in that mode include:
- POSTGRES_HOST
- POSTGRES_PORT
- POSTGRES_DB
- POSTGRES_USER
- POSTGRES_PASSWORD
- VAULT_ADDR
- VAULT_TOKEN
Start with:
docker compose -f docker-compose.external.yml up -d
Project Structure
Primary runtime files:
- src/index.js: stdio startup
- src/http/index.js: HTTP startup
- src/http/server.js: HTTP transport/auth/rate limiting
- src/mcp/server.js: MCP tool registration and auth checks
- src/services/targetService.js: SlackServiceClient adapter
- src/config/env.js: env parsing/validation
Skeleton infrastructure (preserved):
- src/services/vault.js
- src/services/configStore.js
- docker-compose.yml
- docker-compose.external.yml
- vault-production/
Testing
Current tests include:
- Slack MCP tool behavior and mutating auth requirements
- HTTP transport auth/initialize/health behavior
- Existing skeleton Vault/config infrastructure tests
Run all:
npm test
Known Follow-up Work
The server behavior is Slack-specific, but some legacy skeleton assets remain intentionally preserved (Vault/Postgres flows and migration docs). If desired, those docs and compose defaults can be slimmed to a Slack-only deployment profile in a follow-up pass.
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.
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.
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.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.