PagerDuty MCP Server by PagerDuty

# Fix list: PagerDuty MCP Server

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/pagerduty-mcp, the October 2026 research run, assessed 1 October 2026. Grade D, 48.5 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 PagerDuty MCP Server: 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. Payments & pricing, 10 out of 100, up to 11.3 more on the total

Why it scored 10: No x402, MPP or L402 (0). Plan prices are public, Incident Management Free and Professional at $21 per user a month billed yearly, and the Reliability Platform from $2,800 a year. No per-call price and no statement that MCP costs nothing extra (10). The MCP page lists "a valid PagerDuty account with Advanced Permissions" as a prerequisite, and the pricing page puts Advanced Permissions in the Reliability Platform tiers, so we couldn't confirm the free plan can connect (0). A person signs up and creates a key or an OAuth client (0).

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.

## 2. Reliability, 52 out of 100, up to 9.6 more on the total

Why it scored 52: Scored as a hosted MCP server. status.pagerduty.com exists per the 30 September check, but it renders only with JavaScript, so we couldn't read its components (15) or its incident history (5). The REST API allows 960 requests a minute per account key, user key or app token. The MCP docs don't say whether the same limit applies to mcp.pagerduty.com (10). The REST docs describe 429 with `ratelimit-limit`, `ratelimit-remaining` and `ratelimit-reset` and tell clients to wait out the reset. No `Retry-After`, and nothing MCP-specific (12). No SLA on the pricing page (0). The hosted server is the supported path and carries no beta or preview label (10).

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.

## 3. Agent ergonomics, 49 out of 100, up to 8.3 more on the total

Why it scored 49: The hosted docs describe about 16 tool groups without counts. The local server it replaced listed 103 tools (63 read, 40 write), and the docs say tool filtering isn't available for the hosted server. Scoped OAuth makes calls fail rather than shrinking the list (5). Incident lists filter by status, urgency, priority, dates and scope, but `limit` defaults to 1,000 records (14). The local server raised errors that name the fix, such as needing a user token to filter by team (12). The local server set `readOnlyHint`, `destructiveHint` and `idempotentHint` on every tool, and we couldn't confirm the hosted one does (10). Few required parameters (8).

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.

## 4. Schema & documentation, 55 out of 100, up to 7.3 more on the total

Why it scored 55: The hosted tool list and schemas aren't published. The docs say to call tools/list, and point at an MCP API reference on the developer portal we didn't read. The archived local server it replaced had typed Pydantic inputs, flattened in 1.0.0 to remove `$ref` (15). llms.txt and Markdown docs at docs.pagerduty.com (10). The local server's docstrings list each argument and its allowed values in prose, and we can't confirm the hosted text matches (10). Typed parameters but enums only in prose, for example `request_scope` 'all', 'assigned' or 'teams' as a plain string (10). The local repository had a per-domain tool reference with examples, retired with the archive (7). No changelog for the hosted server. The local CHANGELOG had one entry (3).

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.

## 5. Security & auth, 66 out of 100, up to 6 more on the total

Why it scored 66: OAuth with metadata discovery and static client credentials, no dynamic client registration. Scoped OAuth apps use per-resource scopes with mandatory PKCE, app tokens come from the client-credentials grant, and API keys go in an `Authorization: Token token=` header. No secrets in URLs (28). The local server was read-only unless `--enable-write-tools`. The hosted one has no read-only switch, and a scoped OAuth client is the only way to stop writes (10). Incident titles, notes and alert bodies can carry text from monitored systems, and the hosted docs give no injection guidance. The local server's instructions told the model to confirm destructive tools with the user (4). Audit Trail Reporting comes with the Reliability Platform tiers (8). SOC 2 Type II, ISO 27001, FedRAMP Low and a responsible-disclosure page. No security.txt per the 30 September check (16).

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.

## 6. Maintenance & community, 32 out of 100, up to 6 more on the total

Why it scored 32: No hosted changelog. The local server's last version bump was 1.1.0 on 7 July 2026, 86 days before this check (20). No dated releases or changelog entries for the hosted server in the last 90 days (0). Closed service with no public changelog. Support runs through PagerDuty's knowledge base and support desk (5). The official registry entry io.github.PagerDuty/pagerduty-mcp is still version 0.2.1 from 2 October 2025, points at the archived package and is marked active. The hosted server isn't registered (5). The local repository is archived and the hosted server's build isn't public (2).

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.

## 7. Transparency & trust, 64 out of 100, up to 3.2 more on the total

Made of editorial 48, provenance 80.

Why it scored 64: The hosted server is closed with published terms, and the archived local server was Apache-2.0 (18). Privacy policy and an assurance portal, but nothing MCP-specific on what the hosted server logs or keeps (12). The local server's deprecation was announced in the README and archived on 4 September 2026 with a pointer to the hosted server, but the notice and the archive came together, with no lead time, and the GitHub Pages docs had been retired on 25 August (10). US and EU endpoints documented. We didn't read a subprocessor list (8).

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):

- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 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 hosted server's tool list, schemas and annotations, since they're only available through tools/list with an account
- unchecked: status.pagerduty.com components and incident history, because the page renders only with JavaScript
- unchecked: whether the free Incident Management plan can connect, given the Advanced Permissions prerequisite
- We couldn't read support.pagerduty.com/main/docs/pagerduty-mcp-server because the fetch proxy rate-limited us and told us not to retry it. We used the docs.pagerduty.com copy instead
- The listing's lastRelease of 2026-09-04 is the archive date, not a release. The local server's last version bump was 1.1.0 on 7 July 2026 and we didn't confirm its PyPI date
- The listing's docsUrl (pagerduty.github.io) was retired on 25 August 2026, so we patched it

## Weaknesses

- No read-only mode on the hosted server, where the local one was read-only by default
- Hosted tool list, schemas and changelog aren't published
- No native tool filtering, so the full list loads whatever the OAuth scopes allow
- Advanced Permissions, a Reliability Platform feature from $2,800 a year, is listed as a prerequisite
- The official registry entry still points at the archived 0.2.1 package

## 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.

- Use a scoped OAuth client with read scopes only if the agent shouldn't acknowledge, resolve or edit anything
- Use a user API key or user OAuth for `assigned` or `teams` filters. Account-level keys fail those calls
- Pass `limit` and a `since` date on incident lists. The default limit is 1,000 records
- EU accounts must use mcp.eu.pagerduty.com
- On 429, wait `ratelimit-reset` seconds. There's no Retry-After header

## What the review panel asked for

- dated hosted changelog
- hosted read-only mode
- publish the hosted tool list
- add a hosted changelog

## 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.