slack-messages

slack-messages

Enables fuzzy searching and browsing of Slack messages, users, and channels via CLI or MCP server integration.

Category
Visit Server

README

slack-messages

Fuzzy search and browse Slack messages from the command line or as an MCP server.

Features

  • Fuzzy search with typo tolerance across all your Slack messages
  • Browse recent messages, users, and channels
  • User name resolution - shows display names instead of user IDs
  • Context display - see messages before/after each match
  • Filter by sender or date range
  • Thread support - indexes thread replies alongside parent messages
  • Incremental updates - quickly fetch only new messages
  • Multiple interfaces - CLI or MCP server

Requirements

  • Node.js 22+
  • Slack user token with appropriate scopes (see below)

Installation

npm

npm install -g @cardmagic/slack-messages

From source

git clone https://github.com/cardmagic/slack-messages.git
cd slack-messages
pnpm install
pnpm build
npm link

Getting a Slack Token

You need a User OAuth Token (starts with xoxp-), NOT a Bot token.

Important: Bot tokens (xoxb-...) only see the bot's own empty DMs. You must use a User token (xoxp-...) to access your messages.

Required Scopes

Add these under "User Token Scopes" (not "Bot Token Scopes"):

Scope Purpose
channels:history Read public channel messages
groups:history Read private channel messages
im:history Read DM messages
mpim:history Read group DM messages
users:read Get user display names
channels:read List public channels
groups:read List private channels
im:read List DMs
mpim:read List group DMs

Step-by-Step: Create a Slack App

  1. Go to the Slack API portal

  2. Create a new app

    • Click "Create New App"
    • Choose "From scratch"
    • Enter a name (e.g., "Message Search")
    • Select your workspace
    • Click "Create App"
  3. Add User Token Scopes (this is critical!)

    • In the left sidebar, click "OAuth & Permissions"
    • Scroll down to find "User Token Scopes"
    • ⚠️ NOT "Bot Token Scopes" - that's a different section!
    • Click "Add an OAuth Scope" and add ALL of these:
      • channels:history
      • channels:read
      • groups:history
      • groups:read
      • im:history
      • im:read
      • mpim:history
      • mpim:read
      • users:read
  4. Install the app to your workspace

    • Scroll back up to "OAuth Tokens for Your Workspace"
    • Click "Install to Workspace"
    • Review the permissions and click "Allow"
  5. Copy the correct token

    • After installing, you'll see TWO tokens on the OAuth page:
      • "User OAuth Token" - starts with xoxp- - USE THIS ONE
      • ❌ "Bot User OAuth Token" - starts with xoxb- - don't use this
    • Copy the xoxp- token
  6. Add it to slack-messages

    slack-messages auth
    # Paste your token when prompted (input is hidden)
    

Verify Your Token

After adding your token, verify it's correct:

slack-messages index

If you see "Messages: 0" but you know you have messages, you likely used a Bot token by mistake. Go back to step 5 and get the User OAuth Token.

Security Notes

  • Your token is stored locally in ~/.slack-messages/config.json
  • Never commit or share your token
  • You can revoke the token anytime from api.slack.com/apps

Usage

Initial Setup

# Add your Slack workspace (token is prompted securely)
slack-messages auth

# Build the search index (fetches all messages)
slack-messages index

# For incremental updates later
slack-messages index --update

CLI Commands

Browse Commands

# Show most recent messages
slack-messages recent

# List users by recent activity
slack-messages contacts --limit 10

# List channels/DMs with message counts
slack-messages conversations

# Show recent messages from someone
slack-messages from "John Smith"

# Show full conversation in a channel
slack-messages thread "general" --after 2024-12-01

Search Commands

# Search for messages
slack-messages search "quarterly report"

# Filter by sender
slack-messages search "project update" --from "Sarah"

# Filter by date
slack-messages search "meeting" --after 2024-01-01

# Adjust result count and context
slack-messages search "deadline" --limit 20 --context 5

# Show index statistics
slack-messages stats

Workspace Management

# List configured workspaces
slack-messages workspaces

# Remove a workspace
slack-messages remove T0123456789

Search Options

Option Description
-f, --from <sender> Filter by sender name
-a, --after <date> Only messages after date (YYYY-MM-DD)
-l, --limit <n> Max results (default: 10)
-c, --context <n> Messages before/after (default: 2)

MCP Server

Run as an MCP server for Claude Code integration:

slack-messages --mcp
# or
slack-messages mcp

Add to your Claude Code configuration:

claude mcp add --transport stdio slack-messages -- slack-messages --mcp

Or manually in your MCP config:

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

Available MCP Tools:

Tool Description
search_messages Search messages with fuzzy matching
recent_messages Get most recent messages
list_contacts List users by activity
list_conversations List channels/DMs with counts
get_thread Get conversation in a channel
get_message_stats Get index statistics

How It Works

  1. Authentication: Store your Slack user token locally
  2. Indexing: Fetch messages via Slack API and build local indexes:
    • Lists all conversations (channels, DMs, group DMs)
    • Fetches message history for each conversation
    • Fetches thread replies for messages with replies
    • Caches user information for name resolution
  3. Storage: Index files in ~/.slack-messages/:
    • config.json - Workspace tokens
    • index.db - SQLite FTS5 database
    • fuzzy.json - MiniSearch index for typo tolerance
    • stats.json - Index statistics and cursors for incremental updates
    • users.json - Cached user information

Rate Limiting

The Slack API has rate limits. The tool uses the official @slack/web-api client which handles retries automatically. For large workspaces, initial indexing may take a while.

License

MIT

Recommended Servers

playwright-mcp

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.

Official
Featured
TypeScript
Magic Component Platform (MCP)

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.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

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.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

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.

Official
Featured
TypeScript
Kagi MCP Server

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.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

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.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured