BambooHR

# Fix list: BambooHR

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/bamboohr, the October 2026 research run, assessed 7 October 2026. Grade C, 61.7 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 BambooHR: 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, 50 out of 100, up to 10 more on the total

Why it scored 50: Graded on the REST API, which is generally available. The MCP server is marked beta. Status.io page at status.bamboohr.com with four components, three data centre regions and a filterable history (20). Since 9 July 2026 it records three incidents. Access and login problems in all regions for 75 minutes on 8 September, slowness and login errors in the Ireland region for about 1 hour 50 minutes on 5 October, and delayed background processing for about 10 hours on 27 July. We counted 8 September as one major outage (10). No rate limit numbers found. The technical overview says requests can be throttled and the terms reserve the right to do so (0). Rate-limited responses carry `Retry-After`, with 429 replacing 503 from 16 September 2026, but `Idempotency-Key` appears on one operation in 389 (10). The terms commit to availability 24 hours a day with no percentage or remedy, which we didn't count as an SLA (0). REST API is 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.

## 2. Payments & pricing, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). Plan prices are public per employee per month (Core $10, Pro $17, Elite $25, or flat $250, $425 and $650 a month for 25 employees or fewer). Volume discounts aren't quantified and the pricing page doesn't say which plans include the API, so we scored it between plan-only and per-unit (15). The free trial page says no credit card is required, and the developer portal account is free (20). Signup, API key creation and OAuth consent all happen in a browser (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.

## 3. Agent ergonomics, 61 out of 100, up to 6.3 more on the total

Why it scored 61: The API sizes responses with `fields`, `select` and page limits on newer endpoints. The MCP server loads 56 tools with no toolsets, though a Claude user can switch tools off in the connector (18). `list-employees` has cursor paging, `filter` and `sort`, and newer lists take `page` and `pageSize` up to 100. Older endpoints such as the directory return everything at once (16). Newer endpoints answer with problem+json. Permission gaps can return 200 with rows or fields silently missing, which the MCP page documents (14). `Idempotency-Key` on one operation, several deletes documented as idempotent, and MCP tool annotations couldn't be read without an account (6). One maintained official SDK, for PHP, with built-in retries. The .NET and Java SDKs are marked unmaintained (7).

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.

## 4. Security & auth, 64 out of 100, up to 6.3 more on the total

Why it scored 64: OAuth 2.0 authorisation code flow with about 200 scopes in read and `.write` pairs, one-hour access tokens and refresh through `offline_access`. The alternative is a per-user API key over Basic auth that carries all of that user's permissions. PKCE wasn't found in the docs (26). Read scopes without `.write`, permissions that follow the user's access level, and an AI Connectors app that an admin must enable and can limit to chosen access levels. Disabling it revokes every connection. No server-side confirmation for writes (15). The API returns text written by employees and applicants. The connector page advises limiting tools, reviewing writes and keeping sessions to BambooHR, without naming prompt injection (8). Webhook delivery logs for 14 days. No log of API or MCP calls found in the docs (2). The security page lists annual SOC 1 and SOC 2 audits, third-party penetration tests and a bug bounty, with no public intake address found. No security.txt (13).

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.

## 5. Maintenance & community, 62 out of 100, up to 3.3 more on the total

Why it scored 62: The newest dated changelog entry is 26 August 2026, 42 days before the check. A status code change was scheduled for 16 September (20). 14 dated entries between 23 July and 26 August (20). Closed service with a dated changelog, a support address and a feedback form. No public developer forum found (9). The PHP SDK is current to changelog 2.0.2 of 9 July 2026, with its last tag v2.0.1 from December 2025, and no other maintained SDK. BambooHR's MCP server isn't in the official MCP registry (6). The SDK repository has generation workflows, PHPStan and PHPUnit configuration, and a last commit on 31 August 2026 (7).

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.

## 6. Transparency & trust, 74 out of 100, up to 2.3 more on the total

Made of editorial 64, provenance 83.

Why it scored 74: Closed service with terms of service updated 18 September 2026 and developer terms updated February 2026. The PHP SDK is MIT (17). Privacy notice dated September 2025 with a section on customer data, and a DPA last updated 3 June 2022. Customer data is available for 30 days after a subscription ends and then deleted. The terms say payroll records are kept up to seven years and the privacy notice says at least ten (17). The changelog states that a deprecated label carries no removal date unless a separate notice gives one, and the move from 503 to 429 was announced with a date. The developer terms promise notice of breaking changes only where practicable (12). A subcontractor list updated July 2026 names each vendor, its service and location, with LLM providers Anthropic, Cohere and OpenAI, and the security page names hosting regions in the United States, Canada and Ireland (18).

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):

- Terms of service: read, states 6 of the 7 things a reader expects, and has 3 clauses that cost points (3.1 of 10)
- security.txt: not found (0 of 10)

## 7. Schema & documentation, 88 out of 100, up to 2 more on the total

Why it scored 88: OpenAPI 3.1 spec at openapi.bamboohr.io with 277 paths and 389 operations (25). llms.txt on the docs site, and each docs page is served as Markdown at its `.md` URL (10). 376 of 389 operations carry a description longer than 80 characters, and many name the operation to use instead and how permissions change the response (18). 240 enums, numeric limits and required fields in the spec. Older v1 endpoints still take comma-separated field lists and XML bodies (11). Examples throughout and problem+json errors with `code` and `fields` on newer endpoints. Older endpoints report errors in an `X-BambooHR-Error-Message` header that the docs say not to parse (11). Path versions v1, v1_1, v1_2 and v2 and a dated changelog, though the spec's own version stays at 1.0 (13).

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.

- unchecked: the numeric rate limits. None are published in the docs, spec or terms we read
- unchecked: MCP tool definitions, input schemas and readOnlyHint or destructiveHint annotations, which need a BambooHR account to list
- unchecked: the length of the free trial and whether a trial account can create API keys
- unchecked: which plans include API access. The pricing page doesn't say
- unchecked: whether the switch from 503 to 429 took effect on 16 September 2026 as scheduled
- unchecked: where the bug bounty accepts reports, and the SOC 2 report itself (trust centre needs registration and an NDA)
- Whether the OAuth flow supports PKCE. Not found in the reviewed documentation
- The developer terms give the address as 24 Future Way and the DPA as 42 Future Way, as our reader returned them
- Disclosure. Anthropic is on BambooHR's subcontractor list and BambooHR ships a Claude connector. These grades are written by agents running on Anthropic's Claude models, and the checklist was applied as for any listing

## Weaknesses

- No rate limit numbers in the reviewed documentation. The terms reserve the right to throttle
- The MCP server is in beta, loads 56 tools and has no dynamic client registration
- Restricted records and fields can be dropped from MCP and API results with no marker
- PHP is the only maintained official SDK. The .NET and Java SDKs are marked unmaintained
- No security.txt, and the SOC 2 report sits in a trust centre behind registration and an NDA

## 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.

- Treat a short or empty result as what this caller may see. `list_employees` drops employees when a filter or sort field is restricted
- Honour `Retry-After` on 429. Rate-limited calls returned 503 before 16 September 2026, so handle both
- Ask for read scopes only unless the task writes. Write access needs the matching `.write` scope
- Request `offline_access` to receive a refresh token. Access tokens last one hour
- Use `list-employees` with `fields`, `filter` and cursor paging in place of the unpaginated directory endpoint

## 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.