# Fix list: Apideck Accounting API + MCP From Anchor Terminal's listing at https://www.anchorterminal.com/tools/apideck-accounting, the October 2026 research run, assessed 1 October 2026. Grade BB, 73.2 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 Apideck Accounting 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. Payments & pricing, 30 out of 100, up to 8.8 more on the total Why it scored 30: No x402, MPP or L402 (0). Plan prices by number of consumers are public, from $599 a month for 25 to $4,250 for 500, which the checklist treats as plan-only (10). 30-day trial with full access, and the accounting page says no credit card is required (20). Browser signup, then a Vault session per customer. No programmatic route (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. Security & auth, 66 out of 100, up to 6 more on the total Why it scored 66: One API key per application, regenerable from the dashboard, with no per-key scopes, sent in a header and never in a URL (20). The MCP filters tools to read, write or destructive, can lock the consumer with --lock-identity or ?lock_identity=true, and destructive tool descriptions tell the model to confirm with the user. Data scopes on Scale (17). Ledger records carry text written by third parties (supplier names, invoice notes) and we found no prompt-injection guidance. Identity lock limits what an injected prompt can reach (5). Request and webhook logs in the dashboard, configurable log retention on Enterprise (12). SOC 2 Type 2 claimed on the product page and a responsible-disclosure contact at security@apideck.com. security.txt returns 404 and we found no bug bounty (12). 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. Reliability, 75 out of 100, up to 5 more on the total Why it scored 75: Atlassian Statuspage at status.apideck.com, history readable through its RSS feed (20). One incident since early July, dashboard request and webhook logs delayed on 15 September with API traffic unaffected, so minor and off the API path (25). Apideck publishes no limits of its own, only the connector's limit mirrored in x-downstream-ratelimit-* headers, so no numbers (0). The rate-limit guide says to schedule retries from x-downstream-ratelimit-reset, which Apideck fills from the connector's Retry-After, and MCP write tools say a retry may create duplicates, but there are no idempotency keys on writes (10). An SLA document dated 23 February 2026 in the compliance centre, with the contractual uptime SLA on Enterprise (10). GA (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. ## 4. Transparency & trust, 76 out of 100, up to 2.1 more on the total Made of editorial 61, provenance 90. Why it scored 76: Closed service with dated terms (1 January 2026), and an MIT MCP server whose public repository is a compiled mirror rather than source (18). Privacy policy dated 9 February 2026, a DPA from 10 January 2022 and a subprocessor list from 1 January 2026 in the compliance centre. The privacy policy sits on iubenda, which refused our fetch, so we couldn't read retention periods (15). BREAKING.md lists pending removals with sunset dates and the spec carries x-apideck-sunsetting blocks. None pending today (18). A subprocessor list is published, data locations not found. The local MCP sends PostHog events only when POSTHOG_API_KEY is set (10). 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): - security.txt: not found (0 of 10) ## 5. Agent ergonomics, 88 out of 100, up to 2 more on the total Why it scored 88: 362 tools in all, but the default dynamic mode exposes 4 meta-tools at about 1,300 tokens, with read, write and destructive filters and a code mode. The checklist caps a server over 30 tools at 15; we scored the default an agent gets (22). Cursor pagination, limit up to 200, filter, sort and a fields parameter for field selection (20). Typed errors and a URL elicitation flow when the customer's connection is missing (17). readOnlyHint, destructiveHint and idempotentHint set on every tool from its scope, and write descriptions warn that retries can duplicate. No idempotency keys (15). Consumer id optional with a per-call override, SDKs in TypeScript and Python among others (14). 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, 80 out of 100, up to 1.8 more on the total Why it scored 80: OpenAPI spec v10.60.0 on 1 October, MCP 0.21.0 on 28 September and the TypeScript SDK 0.52.2 on 22 September (30). 74 dated spec releases since 3 July and nine MCP releases since 6 July (20). Closed service with a public changelog and an issue tracker on the MCP mirror, where external pull requests are closed automatically because the code lives in a private generator. We couldn't see issue reply times from git (12). Verified namespace com.apideck/mcp in the official registry and current SDKs, but the registry entry is still 0.1.13 from April against 0.21.0 on npm (13). The public MCP repository holds compiled output only, and its CI is a smoke test plus a container tools/list check (5). 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. Schema & documentation, 92 out of 100, up to 1.3 more on the total Why it scored 92: OpenAPI 3.0.3 specs per unified API in apideck-libraries/openapi-specs (accounting.yml, 208 operations, v10.60.0 on 1 October), which the listing had as missing (25). llms.txt with Markdown twins of the guides (10). MCP tool descriptions give the purpose, say whether a call is read-only, not idempotent or destructive, and say what to do when the connection is missing, but rarely say when to pick another tool (16). Zod schemas from the spec with enums, required fields and limit bounded 1 to 200, though pass_through objects are open (13). Typed error schemas such as TooManyRequestsResponse with type_name ConnectorRateLimitError, and examples in the spec (13). Semver on the spec, a dated CHANGELOG.md and a BREAKING.md for pending removals (15). 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. ## 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. - Apideck's own platform rate limits, if any exist - Retention periods and data locations, since the iubenda privacy policy refused our fetch - Whether Apideck stores synced accounting data or passes it through - Issue reply times on the MCP repository ## Weaknesses - No published rate limits of Apideck's own, only connector limits mirrored in headers - No idempotency keys on writes, so a timed-out create can duplicate - $599 a month for 25 customers after the trial, no free tier - Vendor sandboxes only on Enterprise - Registry entry stuck at 0.1.13 from April while npm is at 0.21.0 ## 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. - Stay in dynamic mode and search for tools. Static mode loads all 362 schemas - Start the MCP with --lock-identity (or ?lock_identity=true) when one agent serves one customer - Read x-downstream-ratelimit-remaining before fanning out, and wait until x-downstream-ratelimit-reset after a 429 - List before you retry a create that timed out. Writes aren't idempotent and there's no idempotency key - Pass fields= on list calls to cut the response to the columns you need ## What the review panel asked for - add an errors page to llms.txt - say when to pick another tool - per-key scopes - publish a security.txt ## 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.