mcp-sentry-custom: A Sentry

You run an MCP server that connects to Sentry to retrieve and analyze issues, stack traces, and debugging data. This server lets you query individual issues or lists of issues from Sentry projects and use those details in conversations or workflows.

How to use

Use an MCP client to connect to this server and request Sentry data. You can fetch a single issue by ID or URL, or pull a list of issues for a specific project. The server returns structured information such as titles, IDs, status, levels, timestamps, event counts, and stack traces. You can also request a preformatted, chat-friendly version of an issue for conversation contexts.

How to install

Prerequisites: you need either Node.js and npm, or Python with pip, depending on how you want to run the server.

Option A: Use uvx (recommended) for direct execution without a local install.

uvx mcp-sentry-custom --auth-token YOUR_SENTRY_TOKEN --project-slug YOUR_PROJECT_SLUG --organization-slug YOUR_ORGANIZATION_SLUG --sentry-url YOUR_SENTRY_URL

Option B: Install via pip and run as a module.

pip install mcp-sentry-custom

uv pip install -e .

python -m mcp_sentry

Option C: Install and run using Docker (if you prefer containers).

docker run -i --rm \
  mcp/sentry \
  --auth-token YOUR_SENTRY_TOKEN \
  --project-slug YOUR_PROJECT_SLUG \
  --organization-slug YOUR_ORGANIZATION_SLUG \
  --sentry-url YOUR_SENTRY_URL

Configuration

Configure your client or orchestration tool to start the MCP server using the commands shown below. Provide your own Sentry token, project slug, organization slug, and Sentry URL where placeholders appear.

Additional configuration examples

{
  "mcpServers": {
    "sentry": {
      "command": "uvx",
      "args": [
        "mcp-sentry-custom",
        "--auth-token", "YOUR_SENTRY_TOKEN",
        "--project-slug", "YOUR_PROJECT_SLUG",
        "--organization-slug", "YOUR_ORGANIZATION_SLUG",
        "--sentry-url", "YOUR_SENTRY_URL"
      ]
    }
  }
}

{
  "mcpServers": {
    "sentry": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "mcp/sentry",
        "--auth-token", "YOUR_SENTRY_TOKEN",
        "--project-slug", "YOUR_PROJECT_SLUG",
        "--organization-slug", "YOUR_ORGANIZATION_SLUG",
        "--sentry-url", "YOUR_SENTRY_URL"
      ]
    }
  }
}

{
  "mcpServers": {
    "sentry": {
      "command": "python",
      "args": [
        "-m", "mcp_sentry",
        "--auth-token", "YOUR_SENTRY_TOKEN",
        "--project-slug", "YOUR_PROJECT_SLUG",
        "--organization-slug", "YOUR_ORGANIZATION_SLUG",
        "--sentry-url", "YOUR_SENTRY_URL"
      ]
    }
  }
}

Debugging

If you need to inspect how the MCP server behaves during development or integration, use the MCP inspector to verify requests and responses.

npx @modelcontextprotocol/inspector uvx mcp-sentry-custom --auth-token YOUR_SENTRY_TOKEN --project-slug YOUR_PROJECT_SLUG --organization-slug YOUR_ORGANIZATION_SLUG --sentry-url YOUR_SENTRY_URL

Notes and troubleshooting

If you encounter authentication or URL errors, double-check that your token, project slug, organization slug, and Sentry URL are correct and have the necessary permissions to access the specified project.

Available tools

get_sentry_issue

Retrieve and analyze a specific Sentry issue by its ID or URL, returning details such as title, issue ID, status, level, first/last seen timestamps, event count, and full stack trace.

get_list_issues

Retrieve and analyze a list of Sentry issues for a specific project, returning details like title, issue ID, status, level, first/last seen timestamps, event count, and basic information.

sentry-issue

Provide formatted issue details from Sentry for use in conversation context.