Nevermined API + MCP by Nevermined

# Fix list: Nevermined API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/nevermined, the October 2026 research run, assessed 1 October 2026. Grade BB, 71.1 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 Nevermined API + 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. Security & auth, 63 out of 100, up to 6.5 more on the total

Why it scored 63: One Bearer API key per environment with no scopes found, OAuth 2.1 with PKCE for browser MCP clients, and a device flow for headless agents (22). The preferred key flow returns the API key in a localhost redirect's query string; it never leaves the machine, so we deduct 5 rather than 10 (17). Delegations cap lifetime spend in cents, number of charges and duration, are revocable with one DELETE call, and each card carries a default $10.00 ceiling across delegations, with Visa passkey binding. `pay_service` and `route_by_intent` with `autoPay` move money with no per-call confirmation (16). `pay_service` returns third-party vendor responses and we found no prompt-injection guidance (5). A payment ledger through `list_payments` and `payment_summary`, plus dashboards (13). Claims SOC 2 Type II, ISO/IEC 27001:2022 and PCI SAQ-D, with reports under NDA through trustcenter.nevermined.ai. No security.txt per the 30 September check, and no disclosure policy or bug bounty found (12). The docs don't state who holds funds; card payments settle to the seller's own Stripe or Braintree account, and we found no money-transmission licence.

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.

## 2. Reliability, 72 out of 100, up to 5.6 more on the total

Why it scored 72: Own status page at status.nevermined.ai, built on Upptime, with per-site history kept in the public nevermined-io/uptime repo (20). The only downtime since 1 July is two 502s on the docs site on 2 September. The live and sandbox API monitors started on 28 July 2026 and show 100%, but they ping the API root rather than real calls, so we give 25 of 30. No rate-limit numbers in the docs, only device-flow polling at about 60 a minute per IP per the 30 September check (5 of 15). Errors carry a `retryable` flag meaning the call should succeed after a backoff, settlement is idempotent and the Orders API takes an idempotency key, but no Retry-After or 429 behaviour is documented (12 of 15). No SLA on any plan (0). The Router, Catalog MCP and payments API are sold as live products with no beta label we could find (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. Payments & pricing, 65 out of 100, up to 4.4 more on the total

Why it scored 65: Payment platforms and wallets take the highest step that applies on the 40-point protocol line. 40 for x402, MPP or L402 on all their own endpoints, 30 on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, 0 for only a protocol of their own. Nevermined's own API isn't paid over x402 or MPP. Its sellers can take MPP through the Express middleware (opt-in since v1.11.0 on 18 August 2026) and get an x402 paywall that settles Nevermined plan tokens rather than plain onchain transfers, so the merchant step (25 of 40). The Router also pays outside x402 (Base) and MPP (Tempo) services. Fees published without a login, 1% on stablecoin and 2% on card settlement, 2% Router fee, and $250 or $500 plans (20). Free plan at $0 with 20 agents and 10 payment plans; the page doesn't say whether a card is needed (15). The first API key needs a human sign-in ("the one thing an agent can't mint itself"), and only discovery through the Catalog MCP is keyless (5).

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.

## 4. Agent ergonomics, 74 out of 100, up to 4.2 more on the total

Why it scored 74: The Catalog MCP docs list 11 tools, three of them (`list_categories`, `search_services`, `get_service`) usable without a key (15 + 3). `search_services` and `list_payments` take filters, with no documented page-size limits (14). Errors return a stable code, message, remediation hint, docs link and retry flag, and MCP refusals come back as readable results with codes such as `BCK.ROUTER.0003` for a cap breach (19). Settlement is idempotent, but `pay_service` has no per-call confirmation and we couldn't check tool annotations (10). Official TypeScript and Python SDKs, with few required fields on a delegation (13).

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, 54 out of 100, up to 4 more on the total

Made of editorial 37, provenance 71.

Why it scored 54: SDKs are Apache-2.0 and the platform is closed under published terms (18). The privacy policy carries no date, still cites the Privacy Shield framework, names only Google Analytics and Google Cloud Storage, and gives no retention period beyond 14 months for analytics. The Data Management Policy is gated under NDA (10). No deprecation policy found, although older API keys are now rejected with `BCK.ROUTER.0008` (3). Subprocessors are limited to the two Google services and data locations aren't stated (6).

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: nevermined.ai, registered 2018-06-14 (8 years) (11 of 15)
- Endpoint on the vendor's domain: api.live.nevermined.app is not on nevermined.ai (0 of 15)
- security.txt: not found (0 of 10)

## 6. Schema & documentation, 80 out of 100, up to 3.3 more on the total

Why it scored 80: The docs index lists three OpenAPI specs (core, OAuth and organisations), which we didn't open, and the Catalog MCP server's source isn't public, so we couldn't read its JSON Schemas (20 of 25). llms.txt with about 288 entries and Markdown pages (10). The Catalog MCP page says for each tool what it does, whether it needs a key and whether it moves money, and tells agents to quote before paying (15). Delegation inputs are typed fields such as `spendingLimitCents`, `maxTransactions` and `durationSecs` (11). A full `BCK.*` error catalogue with `hint`, `docsUrl` and `retryable` fields, plus quickstart examples (14). SDK changelogs and GitHub releases, a `/api/v1` path, and no changelog for the hosted API or Router (10).

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, 89 out of 100, up to 1 more on the total

Why it scored 89: Catalog MCP 1.51.0 published to the official registry on 29 September 2026, and the TypeScript SDK v1.13.0 on 9 September (30). Five TypeScript SDK releases since 18 August and three Python ones since 28 August (20). 102 commits to the TypeScript SDK since early July and Discord support on the free plan. We didn't read the issue queue (15). Listed as `io.github.nevermined-io/catalog` in the official MCP registry (15). Unit, integration and end-to-end test suites run in CI on every push, with Dependabot (9).

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 three OpenAPI specs listed in the docs index, and the Catalog MCP tool JSON Schemas and annotations (source not public)
- The Catalog MCP docs page listed 11 tools to our reader; the listing says 12
- unchecked: whether the free plan needs a card
- Who holds buyer funds for ERC-4337 delegations, and whether Nevermined holds any money-transmission licence

## Weaknesses

- The x402 paywall settles Nevermined plan tokens, so generic x402 clients need its SDK or API to pay
- The first API key needs a person to sign in, and the preferred flow hands the key back in a URL query string
- No published rate limits or SLA
- Privacy policy undated, still cites Privacy Shield, and names only two Google subprocessors
- `pay_service` moves money with no per-call confirmation

## 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.

- Call `quote_service` before `pay_service`; quotes cost nothing
- Ask the person to open the `setup_delegation` URL once; you can't mint the first key or budget yourself
- Branch on the error `code` and retry only when `retryable` is true
- Create a small, short delegation per task; cards stop at $10.00 across delegations unless the owner raises it
- Use `sandbox:` keys and mcp.sandbox.nevermined.app until the flow works

## What the review panel asked for

- Programmatic first-key route
- per-call approval threshold
- state who holds funds

## 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.