# Fix list: Skyfire API + MCP From Anchor Terminal's listing at https://www.anchorterminal.com/tools/skyfire, the October 2026 research run, assessed 1 October 2026. Grade E, 40.6 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 Skyfire 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. Reliability, 19 out of 100, up to 16.2 more on the total Why it scored 19: No status page found, and status.skyfire.xyz doesn't resolve (0), so no incident history (5). No rate limits in the docs index or the error reference (0). The error reference says to branch on `code` and not to retry `FORBIDDEN` or `NOT_ELIGIBLE`, but has no 429, Retry-After or idempotency guidance for token creation (4 of 15). No SLA (0). Production and sandbox are both live, with no beta 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. ## 2. Security & auth, 41 out of 100, up to 10.3 more on the total Why it scored 41: A `skyfire-api-key` header, with separate buyer, seller and enterprise admin key types so a buyer key can't charge and a seller key can't mint. Rotation and revocation weren't documented in what we read (22). A pay token caps what a seller can charge at its amount, expires in 10 seconds to 24 hours and is bound to one seller service, but we found no buyer-side spending cap on an agent and no confirmation on `create-pay-token` (10). `find-sellers` returns third-party listings with no prompt-injection guidance (6). No audit log or per-call history found (0). Tokens are JWTs verified against a public JWKS, with `jti` for replay checks, and Skyfire runs KYB on buyer platforms and KYC through Persona. No security.txt per the 30 September check, and no disclosure policy, bug bounty or SOC 2 found (3). Funds sit in "a digital wallet set up by Skyfire or our service providers", the terms name no bank, custodian or licence, and credits are non-refundable. 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. ## 3. Payments & pricing, 20 out of 100, up to 10 more on the total Why it scored 20: 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. No x402, MPP or L402. Payment runs on Skyfire's own KYAPay tokens, an open specification but none of the three protocols, so the own-protocol step (0). No price list; the terms say Skyfire doesn't currently charge for buying credits or generating tokens and may add fees with 30 days' notice (10). A sandbox environment exists, and we didn't establish whether signup or the sandbox needs a card (10). A person signs up and is identity-checked through Persona before an agent can pay (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. ## 4. Schema & documentation, 59 out of 100, up to 6.7 more on the total Why it scored 59: We found no public OpenAPI file for Skyfire's own API; the "OpenAPI Specs" page tells sellers how to describe token requirements in theirs. The MCP server's four tools are documented with their inputs, but the server is closed (10 of 25). llms.txt with about 70 entries and Markdown pages (10). The docs explain when to use a kya, pay or kya-pay token, while the MCP tool descriptions are one line each (12). Token type is one of three values, `expiresAt` is bounded to 10 seconds to 24 hours, and amounts and `sellerServiceId` are typed (11). Eight error codes documented with a fixed `code`, `message`, `details` body (12). `/api/v1` in paths and no changelog (4). 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. Agent ergonomics, 61 out of 100, up to 6.3 more on the total Why it scored 61: Four compact MCP tools, `create-kya-token`, `create-pay-token`, `create-kya-payment-token` and `find-sellers` (25). `find-sellers` takes a search term, and we found no page-size or field controls (8). Stable error codes with field-level details on validation errors and a note on which errors not to retry (16). No idempotency key on token creation; sellers dedupe on the token's `jti` (6). One SDK, a Node seller SDK at 0.0.7 whose repository isn't public, and none for buyers (6). 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. ## 6. Maintenance & community, 31 out of 100, up to 6 more on the total Why it scored 31: Seller SDK 0.0.7 published to npm on 4 August 2026, 58 days ago (20). No other release or dated change found in the last 90 days (0). No changelog and no public SDK repository; the KYAPay specification repo was last touched on 2 September (4). One pre-1.0 Node SDK and no MCP registry entry found (5). The SDK repository is private, so we can't see CI (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, 56 out of 100, up to 3.9 more on the total Made of editorial 50, provenance 62. Why it scored 56: Closed service under terms dated 25 June 2025. The KYAPay specification is published under the `Community Specification License` and the seller SDK under MIT (17). Privacy policy dated 25 June 2025 links a DPA, names Persona and Google Analytics, keeps data in US data centres, gives no retention periods, permits training AI models on personal data, and says sellers receive any personal information carried in a token (18). No deprecation policy or notices found (0). A service-providers page lists processors, and data locations are stated (15). 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: skyfire.xyz, registered 2023-11-28 (2 years) (7 of 15) - Status page: not found (0 of 10) - 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: whether signup or the sandbox needs a card - unchecked: the 30 September claim that ACH and wire funding need a paid subscription; the terms we read don't mention it - unchecked: whether Skyfire's MCP server is in the official MCP registry - Who holds wallet balances (bank, custodian or Skyfire) and whether Skyfire holds a money-transmission licence ## Weaknesses - No x402 or MPP; both buyer and seller must be on Skyfire - No status page, changelog, rate limits or SLA - Settlement of small charges can take up to about 51 hours (24-hour token, 24-hour charge window, 3-hour settlement) - Credits are non-refundable, and the terms name no bank or custodian for wallet funds - Privacy policy permits training AI models on personal data ## 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. - Buyer keys create tokens and seller keys charge them; a 403 usually means the wrong key type - Set the token amount to the most the task should spend; the seller can't charge more - Use a kya token when a site only needs to know who you are; it can't be charged - Don't retry `FORBIDDEN` or `NOT_ELIGIBLE`; fix the key or wait for approval first - Rehearse against mcp-sandbox.skyfire.xyz before using mcp.skyfire.xyz ## What the review panel asked for - Say what the sandbox waives - per-agent spending cap - documented key revocation ## 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.