{
  "fixes": {
    "slug": "spendesk",
    "name": "Spendesk API + MCP",
    "listing": "https://www.anchorterminal.com/tools/spendesk",
    "markdown": "# Fix list: Spendesk API + MCP\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/spendesk, the October 2026 research run, assessed 8 October 2026. Grade B, 62.3 out of 100.\n\nThis 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.\n\nFor a coding agent working on Spendesk 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.\n\n## 1. Payments \u0026 pricing, 0 out of 100, up to 12.5 more on the total\n\nWhy it scored 0: Read with the hosted rubric. No x402, MPP or L402 (0). The pricing page describes a fixed monthly fee plus variable fees per transaction and asks for a quote, with no figures (0). No free tier or trial found. Demo credentials are requested from Spendesk together with API access (0). A person at a paying customer creates the key or approves the OAuth connection, and dynamic client registration is open to supported partners only (0).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):\n\nThe published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).\n\n- 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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).\n\nPayment 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.\n\nOpen-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.\n\n## 2. Reliability, 55 out of 100, up to 9 more on the total\n\nWhy it scored 55: Read with the hosted lines and scored on the public REST API and the MCP server on the same host. status.spendesk.com on Atlassian Statuspage has components by product area with incident history, though none for the public API (20). From 10 July to 8 October 2026 the page lists eleven incidents, three marked critical and three major. Platform access and latency failed for about 1 hour 55 minutes on 2 September, sign-in to the application for about 59 minutes on 29 September, and international transfers in the EEA for about 22 hours from 16 September. No entry says whether the API was affected, so we scored it between one major outage and several (5). Limits are published as 1,000 requests a minute per company and credential and 500 concurrent requests (15). Every response carries x-ratelimit-limit, remaining and reset headers and the docs give pacing guidance, but an Idempotency-Key exists only for creating an intake and Retry-After is documented on one endpoint (10). The docs say the team attempts to guarantee 99 per cent availability, which is a target and not a published SLA (0). The read API is generally available, while cards, transactions, invoices, purchase orders, webhooks and MCP write actions are experimental or beta (5).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):\n\nHosted APIs, MCP servers, models and platforms.\n\n- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.\n\nLocal packages, SDKs, frameworks and stdio MCP servers.\n\n- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.\n\nProtocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.\n\n## 3. Agent ergonomics, 62 out of 100, up to 6.2 more on the total\n\nWhy it scored 62: The MCP server has 62 tools, which is the lowest band, but tools/list returns only the tools the user's permissions allow and write permissions start unticked, so a default connection sees at most the 36 read tools. Analysis tools return totals so an agent needn't sum pages (14). Lists paginate with filters, though in three styles (page, cursor and offset), the page-based maximum is 30 and a page past the end returns 404. MCP list tools take pageSize up to 1,000 and fetchAll up to 5,000 with a truncated flag (17). Errors carry a stable code, the field at fault and a correlationId, and a page explains every MCP error. A missing permission is indistinguishable from a missing tool (17). An Idempotency-Key is documented only for creating an intake. Payable updates are refused on a stale `version`, and batch calls report each item. No readOnlyHint or destructiveHint annotations are documented, and we couldn't list the tools to check (8). Few parameters are required. No official SDK was found, only a Postman collection and copy-and-run scripts (6).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):\n\n- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.\n\nModels 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.\n\n## 4. Maintenance \u0026 community, 57 out of 100, up to 3.8 more on the total\n\nWhy it scored 57: The newest changelog entry was created on 29 September 2026, MCP write actions arrived on 25 September, and docs pages were updated on 1 October (30). The changelog has ten entries, all created on 28 and 29 September 2026, so they meet the count but look like a backfill, and we gave partial credit (15). Closed service with a public changelog and a support address for API consumers (support-public-api@spendesk.com). No public forum or issue tracker was found (8). No official SDK on npm or PyPI, and no entry in the official MCP registry (0). Nothing is packaged to assess. Convention pages state they are replayed against the demo environment weekly, last on 29 September 2026 (4).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):\n\n- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.\n\nModels are read for deprecation notice periods and model churn rather than release counts.\n\n## 5. Security \u0026 auth, 86 out of 100, up to 2.5 more on the total\n\nWhy it scored 86: OAuth 2.0 authorisation code with PKCE (S256), 37 documented scopes, one-hour access tokens and rotating refresh tokens for MCP and partner apps. API keys carry only the scopes ticked, expire within a year, and a token can request fewer scopes than its key (30). Read and write are separate scopes. MCP write permissions are off on every connection until an Admin enables them with a second factor, each user then ticks them, and every call checks the user's role. Confirmation before a write is an instruction to the assistant, not enforced, and production has no dry-run mode apart from export previews (17). The safety page names prompt injection through supplier names, invoice lines and documents, and advises read-only use and approval of each action (12). An action log per MCP connection records date, tool, status, user, company and correlation ID, without arguments. No equivalent for API keys was found (12). security.txt is valid until 31 December 2026 and describes a bug bounty paying up to 2,000 euros. The security page states ISO 27001:2022. No SOC 2 report or public advisories were found, and the trust centre needs JavaScript (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-security):\n\n- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.\n\nModels 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.\n\n## 6. Schema \u0026 documentation, 87 out of 100, up to 2.1 more on the total\n\nWhy it scored 87: A public OpenAPI 3.1 definition with 106 operations at developer.spendesk.com/openapi/spendesk-public-api.json (25). llms.txt indexes every page and each page has a Markdown copy at the same URL with .md (10). 98 of 106 operations carry a description longer than 80 characters, and several say which endpoint to use instead, such as successful against failed transactions. The MCP tool definitions can't be listed without a customer sign-in, and the public tool reference gives a name, a one-line label, the permission and whether the action can be undone (15). The definition has 270 enums, 55 patterns and required lists throughout (13). 400, 401, 403, 429 and 500 are declared on 101 operations with one error model, and 29 operations carry examples (12). Versioning is by path with Deprecation and Sunset headers. The public changelog has ten entries, all created on 28 and 29 September 2026, and the definition's version field is a fixed 1.0 (12).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels 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.\n\n## 7. Transparency \u0026 trust, 80 out of 100, up to 1.8 more on the total\n\nMade of editorial 64, provenance 96.\n\nWhy it scored 80: Closed service. The terms page lists the customer terms, a DORA addendum, a Spendesk API agreement and partner terms, with a subscription for updates and previous versions. We couldn't open the documents, whose links are drawn by JavaScript (12). The privacy policy (March 2025, version 0.5) names a DPO, EU hosting and retention periods for the data Spendesk controls, and refers to a DPA for platform data, which we didn't read. The MCP docs say that data read by an assistant goes to the assistant's provider (20). Deprecation and Sunset headers carry dates and the policy supports the current and previous version, with no minimum notice period stated. Experimental endpoints may change (12). The sub-processor list dated 31 August 2026 gives each processor's address, purpose, data types, location and safeguards (20).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):\n\n- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).\n\nThe other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.\n\nProvenance checks not met in full (half of this category, computed from checked facts):\n\n- Terms of service: published, but our reader couldn't read it (7 of 10)\n- Privacy policy: read, states 6 of the 8 things a reader expects (8.5 of 10)\n\n## What we couldn't check\n\nWhat 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.\n\n- unchecked: the customer terms, DORA addendum and Spendesk API agreement, whose links on the terms page are drawn by JavaScript\n- unchecked: the trust centre at trust.spendesk.com (JavaScript only), so SOC 2 status and the ISO certificate weren't confirmed there\n- unchecked: MCP tool input schemas and readOnlyHint or destructiveHint annotations, which need a customer sign-in to list\n- unchecked: the DPA and its retention terms for platform data\n- Whether the September 2026 platform incidents affected public-api.spendesk.com. The status page has no API component\n- Whether API access is included on every plan or sold separately. The pricing page lists it in the base package, the docs say to ask a representative\n- Whether the ten changelog entries created on 28 and 29 September 2026 are a backfill of older changes\n- No search for security incidents was made beyond the vendor's own pages\n\n## Weaknesses\n\n- No public price, free tier or self-serve signup. API access and demo credentials are requested from a Spendesk representative\n- Cards, transactions, invoices, purchase orders, webhooks and most writes sit behind experimental scopes granted on request, and may change\n- No official SDK found on npm or PyPI, and the MCP server isn't in the official MCP registry\n- status.spendesk.com has no API component and lists three critical and three major incidents opened between 2 and 29 September 2026\n- An Idempotency-Key is documented only for creating an intake. A repeated purchase order or supplier request creates a second record\n\n## What costs an agent a turn today\n\nThe notes we give agents before they call it. Each one is a workaround an agent shouldn't need.\n\n- Request a token at POST /v1/auth/token with HTTP Basic (client ID and secret). It lasts 3,600 seconds, so renew on a 401\n- Stop paging at the last page calculated from `total` and `pageSize` (maximum 30). A page past the end returns 404, not an empty list\n- With an organisation-level token, send `X-Company-Id` on every v1 call or expect a 400\n- The MCP server refuses API keys. Connect with OAuth authorisation code and PKCE, and treat `Tool not found` (-32601) as a missing permission\n- After an unclear write result, read the object again before retrying. Creating a purchase order or supplier twice creates two\n\n## When it's done\n\nSend 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.\n",
    "grade": "B",
    "score": 62.3,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 0,
        "maxGain": 12.5,
        "reason": "Read with the hosted rubric. No x402, MPP or L402 (0). The pricing page describes a fixed monthly fee plus variable fees per transaction and asks for a quote, with no figures (0). No free tier or trial found. Demo credentials are requested from Spendesk together with API access (0). A person at a paying customer creates the key or approves the OAuth connection, and dynamic client registration is open to supported partners only (0).",
        "checklist": [
          "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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 55,
        "maxGain": 9,
        "reason": "Read with the hosted lines and scored on the public REST API and the MCP server on the same host. status.spendesk.com on Atlassian Statuspage has components by product area with incident history, though none for the public API (20). From 10 July to 8 October 2026 the page lists eleven incidents, three marked critical and three major. Platform access and latency failed for about 1 hour 55 minutes on 2 September, sign-in to the application for about 59 minutes on 29 September, and international transfers in the EEA for about 22 hours from 16 September. No entry says whether the API was affected, so we scored it between one major outage and several (5). Limits are published as 1,000 requests a minute per company and credential and 500 concurrent requests (15). Every response carries x-ratelimit-limit, remaining and reset headers and the docs give pacing guidance, but an Idempotency-Key exists only for creating an intake and Retry-After is documented on one endpoint (10). The docs say the team attempts to guarantee 99 per cent availability, which is a target and not a published SLA (0). The read API is generally available, while cards, transactions, invoices, purchase orders, webhooks and MCP write actions are experimental or beta (5).",
        "checklist": [
          "Hosted APIs, MCP servers, models and platforms.",
          "- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 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.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-reliability"
      },
      {
        "key": "ergonomics",
        "name": "Agent ergonomics",
        "score": 62,
        "maxGain": 6.2,
        "reason": "The MCP server has 62 tools, which is the lowest band, but tools/list returns only the tools the user's permissions allow and write permissions start unticked, so a default connection sees at most the 36 read tools. Analysis tools return totals so an agent needn't sum pages (14). Lists paginate with filters, though in three styles (page, cursor and offset), the page-based maximum is 30 and a page past the end returns 404. MCP list tools take pageSize up to 1,000 and fetchAll up to 5,000 with a truncated flag (17). Errors carry a stable code, the field at fault and a correlationId, and a page explains every MCP error. A missing permission is indistinguishable from a missing tool (17). An Idempotency-Key is documented only for creating an intake. Payable updates are refused on a stale `version`, and batch calls report each item. No readOnlyHint or destructiveHint annotations are documented, and we couldn't list the tools to check (8). Few parameters are required. No official SDK was found, only a Postman collection and copy-and-run scripts (6).",
        "checklist": [
          "- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 57,
        "maxGain": 3.8,
        "reason": "The newest changelog entry was created on 29 September 2026, MCP write actions arrived on 25 September, and docs pages were updated on 1 October (30). The changelog has ten entries, all created on 28 and 29 September 2026, so they meet the count but look like a backfill, and we gave partial credit (15). Closed service with a public changelog and a support address for API consumers (support-public-api@spendesk.com). No public forum or issue tracker was found (8). No official SDK on npm or PyPI, and no entry in the official MCP registry (0). Nothing is packaged to assess. Convention pages state they are replayed against the demo environment weekly, last on 29 September 2026 (4).",
        "checklist": [
          "- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.",
          "Models are read for deprecation notice periods and model churn rather than release counts."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-maintenance"
      },
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 86,
        "maxGain": 2.5,
        "reason": "OAuth 2.0 authorisation code with PKCE (S256), 37 documented scopes, one-hour access tokens and rotating refresh tokens for MCP and partner apps. API keys carry only the scopes ticked, expire within a year, and a token can request fewer scopes than its key (30). Read and write are separate scopes. MCP write permissions are off on every connection until an Admin enables them with a second factor, each user then ticks them, and every call checks the user's role. Confirmation before a write is an instruction to the assistant, not enforced, and production has no dry-run mode apart from export previews (17). The safety page names prompt injection through supplier names, invoice lines and documents, and advises read-only use and approval of each action (12). An action log per MCP connection records date, tool, status, user, company and correlation ID, without arguments. No equivalent for API keys was found (12). security.txt is valid until 31 December 2026 and describes a bug bounty paying up to 2,000 euros. The security page states ISO 27001:2022. No SOC 2 report or public advisories were found, and the trust centre needs JavaScript (15).",
        "checklist": [
          "- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-security"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 87,
        "maxGain": 2.1,
        "reason": "A public OpenAPI 3.1 definition with 106 operations at developer.spendesk.com/openapi/spendesk-public-api.json (25). llms.txt indexes every page and each page has a Markdown copy at the same URL with .md (10). 98 of 106 operations carry a description longer than 80 characters, and several say which endpoint to use instead, such as successful against failed transactions. The MCP tool definitions can't be listed without a customer sign-in, and the public tool reference gives a name, a one-line label, the permission and whether the action can be undone (15). The definition has 270 enums, 55 patterns and required lists throughout (13). 400, 401, 403, 429 and 500 are declared on 101 operations with one error model, and 29 operations carry examples (12). Versioning is by path with Deprecation and Sunset headers. The public changelog has ten entries, all created on 28 and 29 September 2026, and the definition's version field is a fixed 1.0 (12).",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 80,
        "maxGain": 1.8,
        "reason": "Closed service. The terms page lists the customer terms, a DORA addendum, a Spendesk API agreement and partner terms, with a subscription for updates and previous versions. We couldn't open the documents, whose links are drawn by JavaScript (12). The privacy policy (March 2025, version 0.5) names a DPO, EU hosting and retention periods for the data Spendesk controls, and refers to a DPA for platform data, which we didn't read. The MCP docs say that data read by an assistant goes to the assistant's provider (20). Deprecation and Sunset headers carry dates and the policy supports the current and previous version, with no minimum notice period stated. Experimental endpoints may change (12). The sub-processor list dated 31 August 2026 gives each processor's address, purpose, data types, location and safeguards (20).",
        "blend": "editorial 64, provenance 96",
        "checklist": [
          "- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-transparency"
      }
    ],
    "provenance": [
      {
        "label": "Terms of service",
        "value": "published, but our reader couldn't read it",
        "points": 7,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 6 of the 8 things a reader expects",
        "points": 8.5,
        "max": 10
      }
    ],
    "unchecked": [
      "unchecked: the customer terms, DORA addendum and Spendesk API agreement, whose links on the terms page are drawn by JavaScript",
      "unchecked: the trust centre at trust.spendesk.com (JavaScript only), so SOC 2 status and the ISO certificate weren't confirmed there",
      "unchecked: MCP tool input schemas and readOnlyHint or destructiveHint annotations, which need a customer sign-in to list",
      "unchecked: the DPA and its retention terms for platform data",
      "Whether the September 2026 platform incidents affected public-api.spendesk.com. The status page has no API component",
      "Whether API access is included on every plan or sold separately. The pricing page lists it in the base package, the docs say to ask a representative",
      "Whether the ten changelog entries created on 28 and 29 September 2026 are a backfill of older changes",
      "No search for security incidents was made beyond the vendor's own pages"
    ],
    "weaknesses": [
      "No public price, free tier or self-serve signup. API access and demo credentials are requested from a Spendesk representative",
      "Cards, transactions, invoices, purchase orders, webhooks and most writes sit behind experimental scopes granted on request, and may change",
      "No official SDK found on npm or PyPI, and the MCP server isn't in the official MCP registry",
      "status.spendesk.com has no API component and lists three critical and three major incidents opened between 2 and 29 September 2026",
      "An Idempotency-Key is documented only for creating an intake. A repeated purchase order or supplier request creates a second record"
    ],
    "agentNotes": [
      "Request a token at POST /v1/auth/token with HTTP Basic (client ID and secret). It lasts 3,600 seconds, so renew on a 401",
      "Stop paging at the last page calculated from `total` and `pageSize` (maximum 30). A page past the end returns 404, not an empty list",
      "With an organisation-level token, send `X-Company-Id` on every v1 call or expect a 400",
      "The MCP server refuses API keys. Connect with OAuth authorisation code and PKCE, and treat `Tool not found` (-32601) as a missing permission",
      "After an unclear write result, read the object again before retrying. Creating a purchase order or supplier twice creates two"
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-08",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.4",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}