What does this project do?
It sets up a local MCP server that parses Slack export logs and registers custom resources (`search://` and `summary://`) so that Claude Desktop can search Slack data using natural language.
@AdilFayyaz
基本情報
Using Model Context Protocol (MCP) to build a local Slack thread searcher with Claude Desktop and a lightweight Python server.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"Slack-Search": {
"command": "python3",
"args": [
"-m",
"venv",
".venv"
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
A local MCP server that parses Slack export logs and registers custom MCP resources for intelligent search, integrating with Claude Desktop to allow natural language queries of Slack data privately and locally.
Clone the repository, set up a Python virtual environment, install dependencies, and update the Claude Desktop config file (claude_desktop_config.json) with the correct paths to the virtual environment and mcp_server.py. Then open Claude Desktop and ask queries like “Search for onboarding discussions from February” or “Find mentions of ‘launch’ in random.”
search://{query} and summary://{channel}It sets up a local MCP server that parses Slack export logs and registers custom resources (`search://` and `summary://`) so that Claude Desktop can search Slack data using natural language.
The project expects a “Slack-dataset” folder containing Slack export JSON files. A sample dataset is available at https://github.com/preethac/Software-related-Slack-Chats-with-Disentangled-Conversations.
You add the MCP server configuration to Claude Desktop’s config file (`claude_desktop_config.json`), pointing to the local `mcp_server.py` script. Claude Desktop then makes MCP calls to fetch and return contextual results.
Yes. The server runs entirely locally, and no data is uploaded to the cloud. The README emphasizes “all locally and privately” and “without uploading your data to the cloud.”
Python 3, a virtual environment, and the `requirements.txt` packages. The server is started via a command in the Claude Desktop config (e.g., using `/bin/zsh` to activate the venv and run `mcp_server.py`).

Give your AI agent a phone. JooVoice makes real calls for you in 80+ languages, with local accents, across 67 countries, like booking a table, moving an appointment or chasing a refund, and sends back what happened. Call
MCP server for the eSIMfly Business API — let Claude, Cursor and other AI agents search eSIM plans, check usage, diagnose connectivity and (optionally) order
Business email on your own domain for AI agents. Read, search and send mail, manage mailboxes and aliases, and set up DNS automatically, with unlimited mailboxes on one flat plan.

Connect an AI assistant to one Vocenya business account: list calls, leads, bookings and chats, add leads, queue compliant AI callbacks and update Do Not Call.https://github.com/gadyamedia/vocenya-mcp

Give your agent an inbox. Its own address on email, Slack and Telegram. A product of Novu
コメント