Metronome

# Fix list: Metronome

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/metronome, the October 2026 research run, assessed 9 October 2026. Grade B, 69.7 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 Metronome: 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, 39 out of 100, up to 7.6 more on the total

Why it scored 39: Payment platforms take the highest step that applies on the 40-point protocol line. No x402, MPP or L402 was found in the API description, the docs index or the pricing page (0). The Startup plan's prices are public at 0.8 per cent of billing volume and $0.04 per 1,000 ingested events. The Custom plan has no price (18 of 20). The pricing page says "Start free" and the quickstart links a self-serve sandbox sign-up, while production checkout needs a credit card. The sign-up page was not read, so whether the sandbox asks for a card is unconfirmed (15 of 20). A person signs up and creates the token in the web app. The changelog of 10 June 2026 says agents can provision an account through the Stripe Projects CLI, which we did not test and which needs a Stripe account (6 of 20). 39.

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, 68 out of 100, up to 6.4 more on the total

Why it scored 68: Hosted lines. status.metronome.com runs on incident.io with seven components (20). Its feed lists 13 incidents and three maintenance windows between 22 July and 8 October 2026. On 3 September a vendor's partial outage raised API and ingest error rates for two hours. API errors were also elevated on 22 July, 11 September (about 30 minutes) and 8 October (a two-minute burst), and most of the rest were data export or invoice delays. Durations were read on three incident pages only (10 of 30). Default rate limits are published per tier in requests a second (15). 429 responses carry `X-Metronome-Rate-Limit-Type` and the docs advise backoff, `Idempotency-Key` covers every POST and the SDKs retry twice by default. No `Retry-After` was found (13 of 15). The pricing page names enhanced SLAs on the Custom plan and no SLA document was found (0). The API is generally available (10). 68.

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. Security & auth, 68 out of 100, up to 5.6 more on the total

Why it scored 68: Bearer tokens are created in the web app, shown once, separate for sandbox and production, and can be archived. Scoping by access level, environment or endpoint exists but is set by contacting support, and by default a token has its creator's permissions (22 of 30). The API description defines agentic tokens whose writes return `403 AgenticMutationBlocked` and wait at `/v1/approval_requests` for a person. How to create such a token was not found in the docs index. Dashboard roles are documented on a page we did not read (15 of 20). The API returns customer-supplied names, custom fields and event properties, and no prompt-injection guidance was found (5 of 15). `/v1/auditLogs` records the user or token, resource, action, IP address and user agent (14 of 15). SOC 2 Type 2 and SOC 1 Type 2 badges, regular penetration tests per the security page, and a Vanta trust centre we could not read. security.txt returned 404 and no bounty or disclosure policy for the service was found (12 of 20). 68.

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, 75 out of 100, up to 4.1 more on the total

Why it scored 75: Graded on the REST API. List calls take `limit` up to 100 and some switches such as `skip_zero_qty_line_items`. No field selection was found and contract and invoice objects are large (13 of 25). Cursor pagination with `next_page`, and filters on status and dates. Many reads are POST calls with a body (17 of 20). Errors carry an HTTP status and a `message` string with no stable code, and the status-code page gives one remedy per status (11 of 20). `Idempotency-Key` on every POST, `uniqueness_key` on contracts and alerts, and 34-day `transaction_id` deduplication on events (20 of 20). Creating a customer has no required list in the description, ingest aliases stand in for IDs, and official SDKs cover five languages (14 of 15). 75.

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. Transparency & trust, 65 out of 100, up to 3.1 more on the total

Made of editorial 45, provenance 85.

Why it scored 65: The platform is closed. Metronome's legal centre links the Stripe Services Agreement, last modified 28 September 2026, which does not name Metronome. The SDKs are Apache-2.0 (15 of 30). Stripe's privacy policy of 28 April 2026 and the DPA the agreement incorporates cover data handling. The agreement says Stripe need not retain data after the term and grants Stripe a licence to use customer content to develop and improve its services. No Metronome retention period was found beyond 30 days of webhook delivery history (15 of 30). No deprecation policy was found. One operation is marked deprecated, v1 contract reads point new clients to v2, and SDK releases of 16 January and 2 March 2026 removed fields with changelog entries on the day (7 of 20). Stripe's privacy pages link a sub-processors list we did not read, and Metronome's trust centre is script-drawn. No hosting region was found (8 of 20). 45.

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):

- Terms of service: read, states 6 of the 7 things a reader expects, and has 1 clause that costs points (7.1 of 10)
- Privacy policy: read, states 8 of the 8 things a reader expects, and has 1 clause that costs points (8 of 10)
- security.txt: not found (0 of 10)

## 6. Schema & documentation, 88 out of 100, up to 2 more on the total

Why it scored 88: A public OpenAPI 3.0.1 description with 142 operations and 509 schemas (25). llms.txt and a Markdown twin of each docs page (10). 131 of 142 operations carry a description longer than 80 characters and 76 have a section on when to use the call. Few say when not to (17 of 20). The description has 345 enums and 774 required lists, with 36 free-form maps (13 of 15). 118 operations carry a request example and 99 a response example. Errors are one `message` string, and only 87 operations document a 4xx response (10 of 15). Paths are versioned `/v1` and `/v2`, and a dated changelog tags API and SDK entries. No dated API versions (13 of 15). 88.

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.

## 7. Maintenance & community, 82 out of 100, up to 1.6 more on the total

Why it scored 82: The changelog's latest entry is dated 29 September 2026, ten days before the check (30). It has 20 dated entries since 14 July 2026 (20). A support portal and a public changelog exist. The SDK issue queues (4 and 13 open items on the Node.js and Python repositories) were not read (9 of 15). Official SDKs in five languages. Python 4.11.1 was tagged on 28 August 2026 and Node.js 3.10.0 on 23 July, and both repositories were pushed to on 6 October (15). The repositories carry CI with lint, build and test jobs, lockfiles and automated releases. CI results were not read (8 of 10). 82.

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.

- The Stripe Services Agreement governs per Metronome's legal centre but never names Metronome. Whether service-specific terms or an order form add Metronome clauses was not established.
- unchecked: trust.metronome.com is a script-drawn Vanta page, so sub-processors, policies and the full list of certifications were not read.
- unchecked: signup.metronome.com was not read, so whether a sandbox account asks for a card is unconfirmed.
- unchecked: the Stripe Projects CLI route for agents is the changelog's claim of 10 June 2026 and was not tested. No vendor software was run.
- unchecked: how an agentic token is created. The approval operations are in the OpenAPI description and no guide page was found in llms.txt.
- unchecked: the status feed gives no durations. Three incident pages were read, the other ten incidents were scored from their titles.
- unchecked: the MCP registry did not answer a search for metronome. No MCP server was found in the docs index or the Node.js SDK repository.
- unchecked: the RBAC, SSO and webhooks guides, Stripe's DPA and sub-processors list, and the SDK issue queues.
- The 2 March 2026 SDK release removed the contract `priority` field in a minor version. Whether customers had notice before that day was not established, so no deduction was taken.
- The docs tell Python users to run `pip install --pre metronome-sdk` although the repository's latest tag is 4.11.1, a stable version.
- The lead named the docs' LLM token billing as a Stripe listing feature. Metronome has its own domain, API, SDKs, prices and status page, so it is graded apart.

## Weaknesses

- Errors are a single `message` string with no machine-readable code, and the API description documents a 4xx response on 87 of 142 operations
- Token scoping by access level, environment or endpoint is set by contacting support. By default a token has its creator's permissions
- The status feed lists 13 incidents between 22 July and 8 October 2026, among them two hours of elevated API and ingest errors on 3 September
- No SLA document is published. The pricing page names enhanced SLAs on the Custom plan only
- The governing terms are Stripe's Services Agreement, which does not name Metronome and grants Stripe a licence to use customer content to develop and improve its services
- No deprecation policy was found. SDK releases of 16 January and 2 March 2026 removed fields, announced in the changelog on the day

## 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.

- Send an `Idempotency-Key` header on every POST and reuse it on retry. A cached error is returned for the same key, so use a new key after fixing the cause
- Give each usage event a stable `transaction_id`. Metronome ignores repeats for 34 days and accepts at most 100 events per `/v1/ingest` call
- On 429 read `X-Metronome-Rate-Limit-Type` (`client` or `customer`) and back off. Unlisted endpoints default to 8 requests a second
- With an agentic token, expect `403 AgenticMutationBlocked` on writes. Post the original method, path and body to `/v1/approval_requests` and show the person the `approval_url`
- Use a sandbox token for testing. The token alone decides whether a call reaches sandbox or production data

## 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.