# Sentry MCP Server

Connects AI assistants to Sentry for searching errors, analyzing performance, triaging issues, reading documentation, and managing projects — all via the Model Context Protocol.

All connections use OAuth. The first connection will trigger an authentication flow to connect to your Sentry account.

## Connecting

The base MCP server address is: `https://mcp.sentry.dev/mcp`

You can optionally scope the connection to an organization or project:

- `https://mcp.sentry.dev/mcp/{organizationSlug}` — scoped to one organization
- `https://mcp.sentry.dev/mcp/{organizationSlug}/{projectSlug}` — scoped to one project

When scoped, tools automatically default to the constrained org/project and unnecessary discovery tools are hidden. Scoping to a project is recommended when possible.

### Query Parameters

- `?experimental=1` — Enable forward-looking tool variants and experimental features
- `?agent=1` — Agent mode: exposes a single `use_sentry` tool that handles natural language requests via an embedded AI agent (roughly doubles response time)

Parameters can be combined: `https://mcp.sentry.dev/mcp/my-org/my-project?experimental=1`

## Setup Instructions

### Claude Code

```bash
claude mcp add --transport http sentry https://mcp.sentry.dev/mcp/{organizationSlug}/{projectSlug}
```

### Cursor

Use the "Install MCP Server" button, or manually add to MCP settings:

```json
{
  "mcpServers": {
    "sentry": {
      "url": "https://mcp.sentry.dev/mcp/{organizationSlug}/{projectSlug}"
    }
  }
}
```

### VSCode

Command Palette → "MCP: Add Server" → HTTP → enter the endpoint:

```
https://mcp.sentry.dev/mcp/{organizationSlug}/{projectSlug}
```

### Other Clients

Any MCP-compatible client can connect using the HTTP transport at the endpoint URL above.