Skip to content
anavem.com

MCP server

Sentry MCP Server

MCP server published by Sentry that connects AI assistants to Sentry to search errors, analyze performance, triage issues, read documentation and manage projects, as a hosted remote service or a local stdio process.

Maintainer
Sentry (getsentry organization on GitHub; hosted at mcp.sentry.dev)
Licence
FSL-1.1-Apache-2.0 (see limitations)
Last release
0.46.0 (shown as cli@0.46.0 on the tags page), released 2026-10-01; earlier releases 0.42.0 and 0.41.0 on 2026-09-25 (releases page of getsentry/toolkit; the page does not make clear which release belongs to which package)
Last verified Jump to what it can access ↓

What it can access

An MCP server gives an AI application a set of capabilities, and it can do only what its code and the credentials you give it allow. This is what the listing states, based on the sources below. Anavem does not rate it safe or unsafe: check it against what you plan to use it for.

Permissions it asks for

  • Hosted connections use OAuth: "The first connection will trigger an authentication flow to connect to your Sentry account" (mcp.sentry.dev). Connections can be scoped to an organization or project.
  • Stdio transport: README says to create a Sentry User Auth Token with these scopes: org:read, project:read, project:write, team:read, team:write, event:write
  • Capabilities per mcp.sentry.dev: searching errors, analyzing performance, triaging issues, reading documentation, managing projects. The page read does not list individual tools.
  • README: AI-powered search tools (search_errors, search_traces, search_logs, search_issues) need an LLM provider (OpenAI, Azure OpenAI, Anthropic or OpenRouter) configured; without one those tools are unavailable

Data it can reach

The hosted endpoint https://mcp.sentry.dev/mcp is a remote server run by Sentry using the HTTP transport; it connects to your Sentry account after OAuth sign-in and can be scoped to an organization or project. The README also shows passing a Sentry token in an Authorization: Sentry-Bearer header instead. The stdio option runs a local process (npx @sentry/mcp-server) that calls Sentry with your User Auth Token; per the MCP architecture page (2026-10-03) a local STDIO server runs on your machine. For the AI-powered search tools, the README says an LLM provider (OpenAI, Azure OpenAI, Anthropic or OpenRouter) must be configured; it does not describe what data is sent to that provider. The pages read do not include a data-retention statement; check Sentry's own terms and privacy documentation.

How it is installed or connected

Steps from the official page (https://mcp.sentry.dev/) and README (https://github.com/getsentry/toolkit), read 2026-10-03:

  1. Hosted: point an MCP-compatible client at https://mcp.sentry.dev/mcp (HTTP transport) and complete the OAuth sign-in on first connection. The page shows setup instructions for Claude Code, Cursor and VS Code; for Claude Code: claude mcp add --transport http sentry https://mcp.sentry.dev/mcp/{organizationSlug}/{projectSlug}.
  2. Claude Code plugin, as shown in the README:
claude plugin marketplace add getsentry/sentry-mcp
claude plugin install sentry-mcp@sentry-mcp
  1. Stdio, as shown in the README: npx @sentry/mcp-server@latest --access-token=sentry-user-token
  2. Self-hosted Sentry, as shown in the README: add --host=sentry.example.com to the stdio command.

Limitations

  • Licence detail: FSL-1.1-Apache-2.0 (Functional Source License, Version 1.1, Apache 2.0 Future License; LICENSE.md). Per the licence text, permitted purposes exclude a "Competing Use", and a future licence under Apache 2.0 takes effect on the second anniversary of the date the Software is made available.
  • README: the service is "primarily designed for human-in-the-loop coding agents", with tool selection focused on developer workflows and debugging use cases rather than "a general-purpose MCP server for all Sentry functionality". The README also calls the stdio transport "still a work in progress".
  • Requires a Sentry account. The pages read do not state plan requirements.
  • Stdio scopes include write permissions (project:write, team:write, event:write) per the README.
  • The repository appears to have moved: github.com/getsentry/sentry-mcp redirects to getsentry/toolkit, and the install commands still reference getsentry/sentry-mcp. Confirm names before installing.
  • Anavem has not audited this software.

Not sure what to look for? Read what to check before installing.

Quick answers

Who maintains this MCP server?
Sentry (getsentry organization on GitHub; hosted at mcp.sentry.dev).
What can it access?
It asks for: Hosted connections use OAuth: "The first connection will trigger an authentication flow to connect to your Sentry account" (mcp.sentry.dev). Connections can be scoped to an organization or project., Stdio transport: README says to create a Sentry User Auth Token with these scopes: org:read, project:read, project:write, team:read, team:write, event:write, Capabilities per mcp.sentry.dev: searching errors, analyzing performance, triaging issues, reading documentation, managing projects. The page read does not list individual tools., README: AI-powered search tools (search_errors, search_traces, search_logs, search_issues) need an LLM provider (OpenAI, Azure OpenAI, Anthropic or OpenRouter) configured; without one those tools are unavailable. The hosted endpoint https://mcp.sentry.dev/mcp is a remote server run by Sentry using the HTTP transport; it connects to your Sentry account after OAuth sign-in and can be scoped to an organization or project. The README also shows passing a Sentry token in an Authorization: Sentry-Bearer header instead. The stdio option runs a local process (npx @sentry/mcp-server) that calls Sentry with your User Auth Token; per the MCP architecture page (2026-10-03) a local STDIO server runs on your machine. For the AI-powered search tools, the README says an LLM provider (OpenAI, Azure OpenAI, Anthropic or OpenRouter) must be configured; it does not describe what data is sent to that provider. The pages read do not include a data-retention statement; check Sentry's own terms and privacy documentation. We do not label anything safe or unsafe; read the official sources before you install.
What licence does it use?
FSL-1.1-Apache-2.0 (see limitations). Check the terms if you plan to use it commercially.
What are its limitations?
Licence detail: FSL-1.1-Apache-2.0 (Functional Source License, Version 1.1, Apache 2.0 Future License; LICENSE.md). Per the licence text, permitted purposes exclude a "Competing Use", and a future licence under Apache 2.0 takes effect on the second anniversary of the date the Software is made available. README: the service is "primarily designed for human-in-the-loop coding agents", with tool selection focused on developer workflows and debugging use cases rather than "a general-purpose MCP server for all Sentry functionality". The README also calls the stdio transport "still a work in progress". Requires a Sentry account. The pages read do not state plan requirements. Stdio scopes include write permissions (project:write, team:write, event:write) per the README. The repository appears to have moved: github.com/getsentry/sentry-mcp redirects to getsentry/toolkit, and the install commands still reference getsentry/sentry-mcp. Confirm names before installing. Anavem has not audited this software.
When was it last released?
0.46.0 (shown as cli@0.46.0 on the tags page), released 2026-10-01; earlier releases 0.42.0 and 0.41.0 on 2026-09-25 (releases page of getsentry/toolkit; the page does not make clear which release belongs to which package). Verified Oct 3, 2026.

Sources

Other listings in the same category.

All MCP servers →