{
  "fixes": {
    "slug": "bill",
    "name": "BILL",
    "listing": "https://www.anchorterminal.com/tools/bill",
    "markdown": "# Fix list: BILL\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/bill, the October 2026 research run, assessed 8 October 2026. Grade C, 60.9 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 BILL: 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, 25 out of 100, up to 9.4 more on the total\n\nWhy it scored 25: Read with the hosted rubric. No x402, MPP or L402 (0). Plan prices are public, $49, $65 and $89 per user per month for AP and AR and $0 for Spend \u0026 Expense, with API access listed on each. The API has no per-call price, and the per-transaction payment fees aren't on the pricing page we read (10). The sandbox is self-serve, charges no subscription fee and uses a test bank account or test card. Production has a 30-day trial, and whether it needs a payment card wasn't established, so partial credit (15 of 20). A person signs up in a browser and generates the developer key in the web app (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. Security \u0026 auth, 56 out of 100, up to 7.7 more on the total\n\nWhy it scored 56: The AP and AR API signs in with a user's username and password plus a developer key, sent in the request body, and the session carries that user's role with no scopes. The Spend \u0026 Expense API takes a token an ADMIN user generates, with no scopes documented. Sync tokens for app partners can't make payments. The MCP server uses OAuth with PKCE (S256) and a revocation endpoint, with only `openid` and `offline_access` as scopes. No secret travels in a URL (16 of 30). Creating a payment, adding a funding bank account and enabling vendor auto-pay need a session trusted by multi-factor authentication, user roles limit who can pay, bill approval policies are in the API, and the MCP server is read-only. No confirmation step was found for other writes, such as creating a vendor card (16 of 20). Bill and invoice text, vendor names and memos are untrusted content, and no prompt-injection guidance was found (3 of 15). An audit trail endpoint covers vendor creates and edits with the user and source, and users can review and disconnect MCP connections under Connected apps. No audit endpoint for other objects or for API calls was found (7 of 15). Annual SOC 1 and SOC 2 Type II audits and a vulnerability disclosure programme on HackerOne. No bounty is stated and security.txt returns 404 (14 of 20).\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## 3. Agent ergonomics, 60 out of 100, up to 6.5 more on the total\n\nWhy it scored 60: `max` sizes a page, with a default of 20 and a ceiling of 100 on AP and AR lists and 50 on Spend \u0026 Expense lists. No field selection was found (16 of 25). Lists return `nextPage` and `prevPage` cursors and take `filters` with nine operators, including ranges on `createdTime` and `updatedTime`, and `sort` (19 of 20). Errors are an array of objects with a `BDC_` code, severity, category and message, and the common codes page gives a fix for each of 12 codes. Rate limit errors don't state when to retry (15 of 20). No idempotency key on the 203 write operations of the v3 API. `X-Idempotent-Key` is accepted on two webhook subscription calls, and a bulk payment request succeeds or fails as a whole. The MCP server is read-only, and its tool annotations couldn't be read without an account (5 of 20). AP and AR calls need a login first and two headers on every call, and the session ends after 35 minutes idle. No official SDK was found, only a Postman collection (5 of 15).\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, 32 out of 100, up to 6 more on the total\n\nWhy it scored 32: The changelog's latest dated entry is 21 May 2026, 140 days before the check. The docs platform's metadata shows `bill-v3-api.json` updated on 5 October 2026 and the MCP server guide on 6 October, so the API is still changing without dated notes (20 of 30). No dated changelog entry falls in the last 90 days (0). Closed service with a changelog, an RSS feed and a developer programme email address. Response times couldn't be observed, and the changelog has been silent for over four months (7 of 15). No official SDK was found, and the MCP server isn't in the official MCP registry, as its own guide says of connector directories. A Postman collection is published (2 of 15). There is no package to assess. The spec and docs are current (3 of 10).\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. Reliability, 81 out of 100, up to 3.8 more on the total\n\nWhy it scored 81: Read with the hosted lines and scored on the v3 REST API. A Statuspage site at www.billcomstatus.com has 15 components, among them API Servers, Virtual Card and Multi-Factor Authentication (20). Its incident feed shows nothing after 15 April 2026, so the last 90 days are clean (30). Limits are published with numbers, 20,000 requests an hour per developer key, three concurrent per key per organisation, 200 logins an hour and 60 Spend \u0026 Expense calls a minute per token (15). The docs give the two limit errors, `BDC_1144` and `BDC_1322`, and a backoff example in Python. No Retry-After header is documented, and no idempotency key is accepted on the v3 API's write operations, payments included (7 of 15). The docs say production is backed by SLAs, but no SLA document was found and the General Terms of Service supply the service as is (0). The v3 API is generally available. The MCP server is beta (9 of 10).\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## 6. Transparency \u0026 trust, 66 out of 100, up to 3 more on the total\n\nMade of editorial 44, provenance 88.\n\nWhy it scored 66: Closed service with terms at stable URLs, the Developer Terms (effective 3 March 2026) and the General Terms of Service (last updated 10 February 2025) (15 of 30). The Privacy Notice (effective 30 January 2026) covers the products and describes retention by criteria without fixed periods. No data processing agreement was found on the legal index, and no statement on training models with customer data was found in the notice (12 of 30). The Developer Terms commit BILL to commercially reasonable efforts at 30 days' notice of deprecations and breaking changes, and the same document keeps the right to change or discontinue the API at any time. The MFA phone change of 3 April 2026 was announced on 6 February, and the FAQ says there are no plans to deprecate v2 (14 of 20). No sub-processor list or data location statement was found. The legal index has a bank partners page (3 of 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: read, states 6 of the 7 things a reader expects (9.1 of 10)\n- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)\n- security.txt: not found (0 of 10)\n\n## 7. Schema \u0026 documentation, 83 out of 100, up to 2.8 more on the total\n\nWhy it scored 83: Two public OpenAPI 3.0.1 files, `bill-v3-api.json` with 314 operations and `connect-events-api.json` with 12 (25). llms.txt, a Markdown copy of every page and a docs MCP server (10). Every operation has a summary and a description, with a median of 209 characters, links to the guide and notes such as the MFA requirement on payments. Few say when not to call an endpoint (14 of 20). 110 enums, 499 required lists, 271 formats, 82 length limits and 32 patterns. The spec declares no security schemes, so credentials appear as header parameters, and `filters` is a string with its own syntax (11 of 15). 1,025 examples. Errors are documented as one `4XX` and one `5XX` array on 311 operations, with 12 common codes explained and over 500 left to the v2 `Errors.json` call (10 of 15). The path is versioned v3 and the changelog is dated with an RSS feed. Its latest entry is 21 May 2026, and the MCP server, whose guide was updated on 6 October, has no entry (13 of 15).\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## 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 live MCP server's tool list, input schemas and readOnlyHint and destructiveHint annotations, which need a signed-in BILL account\n- unchecked: whether the production 30-day trial needs a payment card, and the per-transaction payment fees, which weren't on the pricing page we read\n- unchecked: the SLA the docs say backs production. No SLA document was found on the developer site or the legal index\n- unchecked: the HTTP status code and headers sent with `BDC_1144` and `BDC_1322`, and whether a Retry-After header is returned\n- unchecked: the help centre article on generating a Spend \u0026 Expense API token, so whether that token can be limited or rotated isn't established\n- unchecked: the HackerOne programme page, which is drawn by script, so its scope and whether it pays bounties weren't read\n- unchecked: API changes after 21 May 2026. The docs platform's metadata shows the v3 spec updated on 5 October 2026, with no changelog entry to say what changed\n- status.bill.com and trust.bill.com didn't answer from our network. The status page is at www.billcomstatus.com\n- The lead named the REST API and Elements. BILL also runs a beta MCP server, and the party to the Developer Terms is Bill.com, LLC, a subsidiary of BILL Holdings, Inc.\n- The General Terms of Service bar automated tools that scrape pages of the website. The clause doesn't name the API, so it is recorded here and not as a weakness\n\n## Weaknesses\n\n- `POST /v3/login` takes a user's username and password with a developer key, and the session carries that user's role with no scopes\n- No idempotency key is accepted on the 203 write operations of the v3 API, payments included. `X-Idempotent-Key` exists only on two webhook subscription calls\n- The MCP server is beta, read-only and limited to US organisations, and MCP clients other than Claude and ChatGPT need BILL's approval by email\n- The changelog's latest entry is dated 21 May 2026, and the MCP server has no entry there\n- No official SDK, sub-processor list, data processing agreement or security.txt was found, and the audit trail endpoint covers vendors only\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- Sign in with `POST /v3/login` and send `sessionId` and `devKey` as headers on every AP and AR call. The session expires after 35 minutes idle\n- Send the `apiToken` header alone on `/v3/spend/` paths. Spend \u0026 Expense calls need no login and are limited to 60 a minute per token\n- Complete the MFA challenge before `POST /v3/payments`. An untrusted session fails with `BDC_1361`\n- Read back payments before retrying a failed `POST /v3/payments`. No idempotency key is accepted, so a blind retry can pay twice\n- Keep to three concurrent requests per developer key per organisation and 20,000 an hour. After `BDC_1144`, wait for the next hour\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": "C",
    "score": 60.9,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 25,
        "maxGain": 9.4,
        "reason": "Read with the hosted rubric. No x402, MPP or L402 (0). Plan prices are public, $49, $65 and $89 per user per month for AP and AR and $0 for Spend \u0026 Expense, with API access listed on each. The API has no per-call price, and the per-transaction payment fees aren't on the pricing page we read (10). The sandbox is self-serve, charges no subscription fee and uses a test bank account or test card. Production has a 30-day trial, and whether it needs a payment card wasn't established, so partial credit (15 of 20). A person signs up in a browser and generates the developer key in the web app (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": "security",
        "name": "Security \u0026 auth",
        "score": 56,
        "maxGain": 7.7,
        "reason": "The AP and AR API signs in with a user's username and password plus a developer key, sent in the request body, and the session carries that user's role with no scopes. The Spend \u0026 Expense API takes a token an ADMIN user generates, with no scopes documented. Sync tokens for app partners can't make payments. The MCP server uses OAuth with PKCE (S256) and a revocation endpoint, with only `openid` and `offline_access` as scopes. No secret travels in a URL (16 of 30). Creating a payment, adding a funding bank account and enabling vendor auto-pay need a session trusted by multi-factor authentication, user roles limit who can pay, bill approval policies are in the API, and the MCP server is read-only. No confirmation step was found for other writes, such as creating a vendor card (16 of 20). Bill and invoice text, vendor names and memos are untrusted content, and no prompt-injection guidance was found (3 of 15). An audit trail endpoint covers vendor creates and edits with the user and source, and users can review and disconnect MCP connections under Connected apps. No audit endpoint for other objects or for API calls was found (7 of 15). Annual SOC 1 and SOC 2 Type II audits and a vulnerability disclosure programme on HackerOne. No bounty is stated and security.txt returns 404 (14 of 20).",
        "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": "ergonomics",
        "name": "Agent ergonomics",
        "score": 60,
        "maxGain": 6.5,
        "reason": "`max` sizes a page, with a default of 20 and a ceiling of 100 on AP and AR lists and 50 on Spend \u0026 Expense lists. No field selection was found (16 of 25). Lists return `nextPage` and `prevPage` cursors and take `filters` with nine operators, including ranges on `createdTime` and `updatedTime`, and `sort` (19 of 20). Errors are an array of objects with a `BDC_` code, severity, category and message, and the common codes page gives a fix for each of 12 codes. Rate limit errors don't state when to retry (15 of 20). No idempotency key on the 203 write operations of the v3 API. `X-Idempotent-Key` is accepted on two webhook subscription calls, and a bulk payment request succeeds or fails as a whole. The MCP server is read-only, and its tool annotations couldn't be read without an account (5 of 20). AP and AR calls need a login first and two headers on every call, and the session ends after 35 minutes idle. No official SDK was found, only a Postman collection (5 of 15).",
        "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": 32,
        "maxGain": 6,
        "reason": "The changelog's latest dated entry is 21 May 2026, 140 days before the check. The docs platform's metadata shows `bill-v3-api.json` updated on 5 October 2026 and the MCP server guide on 6 October, so the API is still changing without dated notes (20 of 30). No dated changelog entry falls in the last 90 days (0). Closed service with a changelog, an RSS feed and a developer programme email address. Response times couldn't be observed, and the changelog has been silent for over four months (7 of 15). No official SDK was found, and the MCP server isn't in the official MCP registry, as its own guide says of connector directories. A Postman collection is published (2 of 15). There is no package to assess. The spec and docs are current (3 of 10).",
        "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": "reliability",
        "name": "Reliability",
        "score": 81,
        "maxGain": 3.8,
        "reason": "Read with the hosted lines and scored on the v3 REST API. A Statuspage site at www.billcomstatus.com has 15 components, among them API Servers, Virtual Card and Multi-Factor Authentication (20). Its incident feed shows nothing after 15 April 2026, so the last 90 days are clean (30). Limits are published with numbers, 20,000 requests an hour per developer key, three concurrent per key per organisation, 200 logins an hour and 60 Spend \u0026 Expense calls a minute per token (15). The docs give the two limit errors, `BDC_1144` and `BDC_1322`, and a backoff example in Python. No Retry-After header is documented, and no idempotency key is accepted on the v3 API's write operations, payments included (7 of 15). The docs say production is backed by SLAs, but no SLA document was found and the General Terms of Service supply the service as is (0). The v3 API is generally available. The MCP server is beta (9 of 10).",
        "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": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 66,
        "maxGain": 3,
        "reason": "Closed service with terms at stable URLs, the Developer Terms (effective 3 March 2026) and the General Terms of Service (last updated 10 February 2025) (15 of 30). The Privacy Notice (effective 30 January 2026) covers the products and describes retention by criteria without fixed periods. No data processing agreement was found on the legal index, and no statement on training models with customer data was found in the notice (12 of 30). The Developer Terms commit BILL to commercially reasonable efforts at 30 days' notice of deprecations and breaking changes, and the same document keeps the right to change or discontinue the API at any time. The MFA phone change of 3 April 2026 was announced on 6 February, and the FAQ says there are no plans to deprecate v2 (14 of 20). No sub-processor list or data location statement was found. The legal index has a bank partners page (3 of 20).",
        "blend": "editorial 44, provenance 88",
        "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"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 83,
        "maxGain": 2.8,
        "reason": "Two public OpenAPI 3.0.1 files, `bill-v3-api.json` with 314 operations and `connect-events-api.json` with 12 (25). llms.txt, a Markdown copy of every page and a docs MCP server (10). Every operation has a summary and a description, with a median of 209 characters, links to the guide and notes such as the MFA requirement on payments. Few say when not to call an endpoint (14 of 20). 110 enums, 499 required lists, 271 formats, 82 length limits and 32 patterns. The spec declares no security schemes, so credentials appear as header parameters, and `filters` is a string with its own syntax (11 of 15). 1,025 examples. Errors are documented as one `4XX` and one `5XX` array on 311 operations, with 12 common codes explained and over 500 left to the v2 `Errors.json` call (10 of 15). The path is versioned v3 and the changelog is dated with an RSS feed. Its latest entry is 21 May 2026, and the MCP server, whose guide was updated on 6 October, has no entry (13 of 15).",
        "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"
      }
    ],
    "provenance": [
      {
        "label": "Terms of service",
        "value": "read, states 6 of the 7 things a reader expects",
        "points": 9.1,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 7 of the 8 things a reader expects",
        "points": 9.3,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "not found",
        "points": 0,
        "max": 10
      }
    ],
    "unchecked": [
      "unchecked: the live MCP server's tool list, input schemas and readOnlyHint and destructiveHint annotations, which need a signed-in BILL account",
      "unchecked: whether the production 30-day trial needs a payment card, and the per-transaction payment fees, which weren't on the pricing page we read",
      "unchecked: the SLA the docs say backs production. No SLA document was found on the developer site or the legal index",
      "unchecked: the HTTP status code and headers sent with `BDC_1144` and `BDC_1322`, and whether a Retry-After header is returned",
      "unchecked: the help centre article on generating a Spend \u0026 Expense API token, so whether that token can be limited or rotated isn't established",
      "unchecked: the HackerOne programme page, which is drawn by script, so its scope and whether it pays bounties weren't read",
      "unchecked: API changes after 21 May 2026. The docs platform's metadata shows the v3 spec updated on 5 October 2026, with no changelog entry to say what changed",
      "status.bill.com and trust.bill.com didn't answer from our network. The status page is at www.billcomstatus.com",
      "The lead named the REST API and Elements. BILL also runs a beta MCP server, and the party to the Developer Terms is Bill.com, LLC, a subsidiary of BILL Holdings, Inc.",
      "The General Terms of Service bar automated tools that scrape pages of the website. The clause doesn't name the API, so it is recorded here and not as a weakness"
    ],
    "weaknesses": [
      "`POST /v3/login` takes a user's username and password with a developer key, and the session carries that user's role with no scopes",
      "No idempotency key is accepted on the 203 write operations of the v3 API, payments included. `X-Idempotent-Key` exists only on two webhook subscription calls",
      "The MCP server is beta, read-only and limited to US organisations, and MCP clients other than Claude and ChatGPT need BILL's approval by email",
      "The changelog's latest entry is dated 21 May 2026, and the MCP server has no entry there",
      "No official SDK, sub-processor list, data processing agreement or security.txt was found, and the audit trail endpoint covers vendors only"
    ],
    "agentNotes": [
      "Sign in with `POST /v3/login` and send `sessionId` and `devKey` as headers on every AP and AR call. The session expires after 35 minutes idle",
      "Send the `apiToken` header alone on `/v3/spend/` paths. Spend \u0026 Expense calls need no login and are limited to 60 a minute per token",
      "Complete the MFA challenge before `POST /v3/payments`. An untrusted session fails with `BDC_1361`",
      "Read back payments before retrying a failed `POST /v3/payments`. No idempotency key is accepted, so a blind retry can pay twice",
      "Keep to three concurrent requests per developer key per organisation and 20,000 an hour. After `BDC_1144`, wait for the next hour"
    ],
    "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"
  }
}