mcp-sentry: A Sentry MCP server

Name: mcp-sentry: A Sentry MCP server
Author: MCP-100

@MCP-100

リポジトリホームページ

mcp-sentry: A Sentry MCP server について

A Model Context Protocol server for retrieving and analyzing issues from Sentry.io

#python #sentry #mcp #mcp-server

基本情報

カテゴリ

開発者ツール

ランタイム

python

トランスポート

stdio

公開者

MCP-100

設定

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": {
    "mcp-sentry": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@qianniuspace/mcp-sentry",
        "--client",
        "claude"
      ]
    }
  }
}

ツール

`issue_id_or_url` (string): Sentry issue ID or URL to analyze

Issue details including:

`project_slug` (string): Sentry project slug to analyze

List of issues with details including:

概要

What is mcp-sentry?

mcp-sentry is a Model Context Protocol server for retrieving and analyzing issues from Sentry.io. It provides tools to inspect error reports, stacktraces, and other debugging information from your Sentry account, aimed at developers using Sentry for error monitoring.

How to use mcp-sentry?

Install via Smithery, uv (uvx), or pip. Configure with your Sentry auth token, project slug, and organization slug in your MCP client (e.g., Claude Desktop or Zed). Invoke the server using the commands specified in the configuration, and use the tools get_sentry_issue and get_list_issues or the prompt sentry-issue to retrieve Sentry issue data.

Key features of mcp-sentry

Retrieve a Sentry issue by ID or URL.
List Sentry issues by project and organization slug.
Return issue details: title, status, level, timestamps, event count.
Provide full stacktrace for an issue.
Includes a prompt for conversational issue retrieval.

Use cases of mcp-sentry

Debugging a specific error by fetching its full stacktrace.
Monitoring recent issues in a project for trend analysis.
Integrating Sentry error data into AI‑assisted debugging workflows.

FAQ from mcp-sentry

What are the runtime requirements?

The server requires Python and can be run via uvx or pip. A Sentry authentication token, project slug, and organization slug are needed for configuration.