
Slack Mcp Server
io.github.stevenvo由 GitHub 账号 stevenvo 发布v0.1.0更新于 Oct 24, 2026
Read Slack conversations, threads, user info, and search messages from your MCP client.
概览
一个本地 MCP 服务器,让助手读取 Slack 工作区的频道消息、话题串、用户资料和搜索结果。
- 可以做什么
- 按时间范围和数量上限读取频道近期消息,并读取包含全部回复和表情回应的话题串。列出你所在的频道,包括公开频道、私有频道、私信和群组私信,并获取频道主题、用途和成员数等元数据。查询用户资料,包括姓名、邮箱、职位、状态和时区。搜索你被提及或参与的会话。为指定消息生成永久链接。
- 适用场景
- 适合让助手跟进项目频道、阅读讨论串、通过搜索找回过去的决定,或在不离开对话客户端的情况下了解人员和频道背景。它只读,因此适合调研和总结,不适合发帖或管理 Slack。
- 运行要求
- 以 Python 包形式在本地运行(需 Python 3.10 或更高版本),可通过 pip 或 uvx 安装。需要一个配置了所列 OAuth 权限范围的 Slack 应用,以及用户令牌(SLACK_USER_TOKEN)或机器人令牌(SLACK_BOT_TOKEN),并需设置 SLACK_WORKSPACE_URL 以生成消息链接。使用机器人令牌时需将机器人邀请进各频道。需要访问 Slack API 的网络连接。
安装
在 SourceWeft 中
- 打开 控制台中的 Slack Mcp Server,将其添加到工作区。
- 为需要使用其工具的对话启用该服务。
Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。
其他 MCP 客户端
参照 仓库 中的启动说明。
README
Slack MCP Server
[PyPI] [Python Version] [License: MIT]
A Model Context Protocol (MCP) server that provides programmatic access to Slack conversations, threads, and workspace information. Designed for AI assistants like Claude to interact with your Slack workspace through a standardized interface.
What is MCP?
The Model Context Protocol is an open standard that enables AI assistants to securely access external data sources and tools. This server implements MCP to give Claude direct access to your Slack workspace.
Features
📨 Message Operations
- Read Channel Messages: Fetch recent messages from any channel with flexible time-based filtering
- Read Thread Messages: Access complete conversation threads including all replies
- Get Message Permalinks: Generate permanent, shareable links to specific messages
🔍 Discovery & Search
- List Your Channels: Enumerate all channels you're a member of (public, private, DMs, group DMs)
- Search Conversations: Find messages where you're mentioned or involved using Slack's search
📊 Metadata & Context
- Get Channel Info: View channel details, topics, purposes, member counts, and settings
- Get User Info: Look up user profiles, emails, titles, timezones, and status information
Quick Start
Option 1: Install from PyPI (Recommended)
The easiest way to use this MCP server is to install it directly from PyPI:
Option 2: Install from Source
After installation, you'll need to set up authentication (see below).
Authentication
Option 1: User Token (Recommended)
A User Token uses your personal Slack permissions and can access all channels you're a member of.
- Go to https://api.slack.com/apps
- Create a new app (or use existing)
- Navigate to "OAuth & Permissions"
- Add the following User Token Scopes:
channels:history- View messages in public channelschannels:read- View basic channel infogroups:history- View messages in private channelsgroups:read- View basic private channel infoim:history- View messages in direct messagesim:read- View basic DM infompim:history- View messages in group DMsmpim:read- View basic group DM infousers:read- View user informationsearch:read- Search messages
- Install the app to your workspace
- Copy the "User OAuth Token" (starts with
xoxp-)
Option 2: Bot Token
A Bot Token is more limited but easier to set up for team-wide access.
- Follow steps 1-2 above
- Add Bot Token Scopes instead (same list but in bot section)
- Install the app
- Copy the "Bot User OAuth Token" (starts with
xoxb-) - Invite the bot to channels you want to read:
/invite @YourBotName
Configure Environment
Create a .env file:
Edit .env and add your token:
Configuration for Claude Code
Add this server to Claude Code by editing your ~/.claude.json file:
Using PyPI Package (Recommended)
Using Local Installation
If you installed from source:
Using Bot Token
If you prefer to use a bot token instead of a user token:
Note: Remember to invite your bot to channels: /invite @YourBotName
Verify Installation
Restart Claude Code and verify the server is running:
You should see slack in the list of active MCP servers.
Available Tools
This server provides 7 MCP tools for interacting with Slack:
📬 read_channel_messages
Retrieve recent messages from any Slack channel with flexible time filtering.
Parameters:
Natural Language Examples:
- "Show me messages from the #general channel in the last 24 hours"
- "Read the last 48 hours of messages from channel C1234567890"
- "Get the most recent 50 messages from C070PDRHQS1"
Returns: Formatted messages with timestamps, user mentions, thread indicators, reactions, and permalinks.
💬 read_thread_messages
Read all messages in a conversation thread, including the parent message and all replies.
Parameters:
Natural Language Examples:
- "Read the full thread with timestamp 1699564800.123456 in channel C1234567890"
- "Show me all replies to message ts 1699564800.123456"
Returns: Complete thread with parent message and all replies, formatted with timestamps and reactions.
📋 get_channel_info
Get detailed metadata about a Slack channel.
Parameters:
Natural Language Examples:
- "Get info about channel C1234567890"
- "Show me details for the #engineering channel"
Returns: Channel name, ID, creation date, member count, privacy status, topic, and purpose.
👤 get_user_info
Look up detailed profile information for any Slack user.
Parameters:
Natural Language Examples:
- "Get info about user U1234567"
- "Show me details for @john.doe"
Returns: User's real name, display name, email, title, status, timezone, and admin/bot flags.
📑 list_my_channels
List all channels you're a member of, with support for different channel types.
Parameters:
Supported Types:
public_channel- Public channelsprivate_channel- Private channelsmpim- Group direct messagesim- Direct messages
Natural Language Examples:
- "List all my channels"
- "Show me all private channels I'm in"
- "List all my DMs"
Returns: Channel names, IDs, member counts, and privacy indicators.
🔍 search_my_conversations
Search for messages across all conversations where you're mentioned or involved.
Parameters:
Natural Language Examples:
- "Search my conversations for 'deployment'"
- "Find messages mentioning 'bug fix' in the last week"
- "Search for messages about 'Q4 planning'"
Returns: Matching messages with channel names, user info, timestamps, and permalinks.
🔗 get_message_permalink
Generate a permanent, shareable link to a specific Slack message.
Parameters:
Natural Language Examples:
- "Get permalink for message 1699564800.123456 in channel C1234567890"
- "Generate a link to this message"
Returns: Permanent URL that works even if the workspace's URL changes.
Usage Examples
Once configured in Claude Code, you can interact with your Slack workspace using natural language. Here are some example queries:
Reading Messages
Following Threads
Discovery & Search
Getting Context
Practical Workflows
Catch up on a project:
Find that important decision:
Get onboarding context:
Finding Channel IDs
Method 1: From Slack URL
When you're in a channel, the URL looks like:
The part after /archives/ is the channel ID: C1234567890
Method 2: Using the MCP Server
Ask Claude:
Method 3: Right-click in Slack
- Right-click on the channel name
- Click "Copy link"
- Extract the channel ID from the URL
Requirements
- Python: 3.10 or higher
- Slack Workspace: Admin access to create and configure a Slack app
- Claude Code: Latest version with MCP support
- Operating System: macOS, Linux, or Windows with WSL
Troubleshooting
"Missing required scopes" Error
Your Slack app needs the proper OAuth scopes configured.
Solution:
- Go to https://api.slack.com/apps and select your app
- Navigate to "OAuth & Permissions"
- Ensure you've added all required scopes listed in the Authentication section
- Reinstall the app to your workspace
- Copy the new token
"Channel not found" Error
For Bot Tokens:
For User Tokens:
- Verify you're a member of the channel
- Check the channel ID is correct (see "Finding Channel IDs" section)
"Not authenticated" Error
Check your configuration:
Common issues:
- Token has expired or been revoked
- Extra spaces or newlines in the token string
- Wrong token type (user vs bot)
- Token not properly quoted in JSON
MCP Server Not Running
Verify server is loaded:
Check logs for errors:
Common solutions:
- Restart Claude Code completely
- Check
~/.claude.jsonsyntax is valid JSON - Verify Python version:
python --version(must be 3.10+) - Test the server manually:
python server.py(if installed from source)
SSL Certificate Errors
If you're behind a corporate proxy (like Zscaler):
Rate Limiting
Slack API has rate limits. If you hit them:
- Reduce the
limitparameter inread_channel_messages - Increase
lookback_hoursto fetch fewer messages - Wait a few minutes before retrying
Rate limit indicators:
Solution: The server will automatically handle rate limits, but you may need to wait.
Security & Privacy
Token Security
- Never commit tokens to version control - The
.envfile is gitignored by default - Revoke compromised tokens immediately at https://api.slack.com/apps
- Use environment variables in production, never hardcode tokens
- Rotate tokens periodically as part of security best practices
Data Access
User Tokens:
- Have access to everything you can see in Slack
- Use your personal permissions
- See all channels you're a member of
- More convenient for personal use
Bot Tokens:
- Limited to channels where the bot is invited
- Separate from personal identity
- Better for team/shared setups
- More granular control
Corporate Environments
If you're using this in a corporate environment:
- Check with your IT/Security team before creating Slack apps
- Be aware of data retention and compliance policies
- Consider using bot tokens for audit trails
- SSL certificate configuration may be required (see Troubleshooting)
Limitations
- Read-only: This server only reads data, it cannot post messages or modify content
- Rate limits: Subject to Slack's API rate limits (Tier 3: ~50 requests per minute)
- Token scope: Can only access channels/conversations the token has permission to see
- Message history: Limited to Slack workspace's message retention policy
Contributing
Contributions are welcome! Here's how you can help:
Reporting Issues
Found a bug or have a feature request?
- Check existing issues at https://github.com/stevenvo/slack-mcp-server/issues
- Create a new issue with:
- Clear description of the problem/feature
- Steps to reproduce (for bugs)
- Expected vs actual behavior
- Your environment (Python version, OS, Claude Code version)
Pull Requests
- Fork the repository
- Create a feature branch:
git checkout -b feature/amazing-feature - Make your changes
- Test thoroughly
- Commit with clear messages:
git commit -m "Add amazing feature" - Push to your fork:
git push origin feature/amazing-feature - Open a Pull Request
Development Setup
Changelog
v0.1.0 (2025-01-06)
- Initial release
- 7 core tools for reading Slack conversations
- Support for user and bot tokens
- Published to PyPI and MCP registry
- Comprehensive documentation
Roadmap
Potential future enhancements:
- Support for Slack Enterprise Grid
- Message filtering by user or reactions
- Export conversations to different formats
- Support for Slack Connect channels
- Caching to reduce API calls
- Webhook support for real-time updates
Suggestions welcome in GitHub Issues!
Related Projects
- Model Context Protocol - The MCP specification
- MCP Registry - Browse more MCP servers
- Slack SDK for Python - The underlying library
- Claude Code - AI assistant with MCP support
Support
- Documentation: You're reading it!
- Issues: https://github.com/stevenvo/slack-mcp-server/issues
- Discussions: https://github.com/stevenvo/slack-mcp-server/discussions
- Slack API Docs: https://api.slack.com/docs
License
MIT License - see LICENSE file for details.
Copyright (c) 2025 Steven Vo
Made with ❤️ for the MCP community
If you find this useful, please ⭐ star the repository on GitHub!
来源:README.md,提交 f11488b
工具
0版本历史
1- v0.1.0最新Sep 16, 2026
评分
已满足适用检查项的 50%
该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。
- 近 30 天内可连通25 分
请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。
- 已声明认证方式10 分
请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。
- 已列出工具15 分
请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。
- 描述必需15 分
- 图标10 分
- 文档15 分
- 许可证10 分