Sentry MCP by Sentry

# Fix list: Sentry MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/sentry-mcp, the October 2026 research run, assessed 1 October 2026. Grade BB, 70.4 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Sentry MCP: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Reliability, 46 out of 100, up to 10.8 more on the total

Why it scored 46: Scored as a hosted MCP server, since mcp.sentry.dev is the default install. status.sentry.io is an Atlassian Statuspage with API, Dashboard and per-region ingestion components, but no MCP component (15). We could read only the incidents on its front page. The JSON feed is blocked by robots rules and the history page didn't render. Those show two EU ingestion delays on 18 September 2026, 3 hours 21 minutes for errors and 40 minutes for spans, both blamed on an upstream network provider (10). The hosted server's limits, 60 calls a minute per user and 300 per IP on /mcp, are in the public Wrangler config, not in user docs (8). Upstream 429s become a typed rate-limit error carrying `retry_after` when Sentry sends it, documented only in the contributor docs (8). No SLA on the pricing page (0). Versions are 0.x and `?experimental=1` variants can change between releases, with no GA statement (5).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 2. Payments & pricing, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). The MCP server has no separate charge and Sentry's plan prices are public, with MCP included on the free Developer plan (20). The Developer plan is free with no card mentioned (20). A person has to create a Sentry account and approve OAuth (0). Self-hosted use of the AI search tools needs your own LLM provider key.

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 3. Security & auth, 75 out of 100, up to 4.4 more on the total

Why it scored 75: The hosted server runs its own OAuth provider with dynamic or metadata-document client registration, and the user picks skills at consent. The upstream Sentry token it holds always asks for `org:read project:write team:write event:write alerts:write`, so the narrowing happens in the MCP layer only (25). Inspect and Seer, both read-only, are the default skills. Triage and project management are opt-in, `?skills=` narrows a direct token, and org or project constraints limit scope. No confirmation step beyond annotations (16). Contributor docs forbid passing untrusted error text to agents and list input validation and region-URL checks, but event messages, breadcrumbs and user reports still reach the model unmarked (9). Sentry's audit log records user actions, and we didn't confirm MCP calls appear there (8). SOC 2 Type I and II, ISO 27001, HIPAA attestation and a disclosure programme at security@sentry.io, no bounty. No advisories for the MCP server (17).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 4. Agent ergonomics, 85 out of 100, up to 2.4 more on the total

Why it scored 85: tools/list shows nine top-level tools, about 27,000 characters (roughly 6,700 tokens), with `search_events` alone at 2,031 characters (18). The other 59 tools load on demand through `search_sentry_tools` and `execute_sentry_tool`, skills narrow the set, and org or project scoping in the URL hides discovery tools (plus 7). Limits on searches and org and project constraints (17). Errors come back as typed, actionable messages, with 404s telling the agent to check the org, project or id (18). 42 tools set `readOnlyHint: true`, 23 set it false, 12 carry `destructiveHint` and 17 `idempotentHint`. But every catalogue tool runs through `execute_sentry_tool`, which is marked destructive as a whole, and issue #1254 (open since 14 August) says this breaks client allowlists and approval prompts (15). Few required parameters, npm only (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 5. Schema & documentation, 87 out of 100, up to 2.1 more on the total

Why it scored 87: Every tool has a Zod input schema published as JSON Schema in toolDefinitions.json (25). llms.txt at docs.sentry.io and linked from mcp.sentry.dev (10). Descriptions are long and structured, with "Use this tool when", `<examples>` and `<hints>` blocks, and some say when not to use the tool, for example `add_issue_note` warns against secrets and unapproved content (18). Inputs carry required lists, `minLength` and `maxLength`, URI formats and nullable unions (13). Examples sit inside the descriptions, and errors map to typed classes with recovery hints (13). GitHub releases and tags. The release guide says to update a CHANGELOG.md the repository doesn't have (8).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 6. Transparency & trust, 83 out of 100, up to 1.5 more on the total

Made of editorial 70, provenance 96.

Why it scored 83: FSL-1.1-Apache-2.0, source-available with a two-year conversion to Apache-2.0, not an OSI licence today (20). Sentry publishes a privacy policy, DPA and a subprocessor list updated 1 June 2026 that names Anthropic and OpenAI as AI subprocessors and Cloudflare, AWS and GCP in the EU and US. TELEMETRY.md shows the hosted server records tool arguments and tool results in Sentry's own monitoring projects, and mcp.sentry.dev doesn't say so (22). The `docs` skill is marked deprecated in its own description and tools have been removed with commit notes, but there's no deprecation policy or dates (8). Subprocessors and their locations are disclosed. The stdio package sends telemetry only when a DSN is set (20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: sentry.dev, registered 2019-02-22 (7 years) (11 of 15)

## 7. Maintenance & community, 90 out of 100, up to 0.9 more on the total

Why it scored 90: 0.42.0 on 2026-09-25, the npm latest (30). 0.38.0, 0.39.0, 0.40.0, 0.41.0 and 0.42.0 since 26 August, and cli@0.46.0 on 1 October (20). 69 open issues and 33 open pull requests. Several feature requests from July and August have no labels, and #1226 (hosted AI search failing with "Feature Unavailable", 4 August) is still open (15). Listed in the official MCP registry as io.github.getsentry/sentry-mcp, published from a workflow (15). Tests, smoke tests and a token-cost check run in CI on every pull request (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: the full 90-day incident history, because status.sentry.io's JSON feed is blocked by robots rules and its history page didn't render for us
- unchecked: whether MCP-initiated writes appear in Sentry's organisation audit log
- We dropped the listing's citation of a community grader's letter grade, since the brief rules out copying a ranking site's score

## Weaknesses

- No MCP component on status.sentry.io and no SLA
- The upstream Sentry token always holds project, team, event and alert write scopes
- `execute_sentry_tool` wraps every catalogue tool, which breaks per-tool approval (issue #1254)
- Functional Source License, not OSI-approved until each version's two-year conversion
- Hosted telemetry records tool arguments and results, and the public site doesn't mention it

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Put the org and project in the URL (mcp.sentry.dev/mcp/<org>/<project>) to drop the discovery tools
- Call `search_sentry_tools` to find a catalogue tool, then `execute_sentry_tool` to run it
- Use `get_sentry_resource` with an issue or trace URL when you have one. It's faster than natural-language search
- Grant Triage only when the task needs to resolve or assign issues. The default skills are read-only
- Expect `?experimental=1` tool variants to change between releases

## What the review panel asked for

- changelog with deprecation dates
- pass inner annotations through
- mark untrusted event text

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.