HiBob

# Fix list: HiBob

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/hibob, the October 2026 research run, assessed 7 October 2026. Grade C, 57 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 HiBob: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 0 out of 100, up to 12.5 more on the total

Why it scored 0: No x402, MPP or L402 (0). Prices are by quote only (0). No free tier or trial found, and the sandbox is a purchased module (0). Credentials come from a customer's admin in the Bob interface, with no programmatic route (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Agent ergonomics, 47 out of 100, up to 8.6 more on the total

Why it scored 47: The `fields` array sizes people responses, but people search returns every matching employee in one response (13). Cursor pagination with a limit up to 200 covers bulk tables, workforce planning and the job catalogue only, and people search filters accept only `root.id` and `root.email` with equals (10). Status codes and recovery advice are documented (no retry on 400, stop on 401 and 403), but fields without permission vanish from a 200 response with no warning (12). No idempotency keys. 304 signals an unchanged record, and the MCP tool annotations could not be read (6). Few required parameters on reads. No official SDK found in the reviewed documentation (6).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 3. Reliability, 67 out of 100, up to 6.6 more on the total

Why it scored 67: Graded on the public REST API. Statuspage at status.hibob.io with 46 components, Public API among them (20). Nine incidents between 9 July and 7 October 2026. Three were major or critical on single modules (employee work tables failing to update for 4 hours 26 minutes on 31 August, Talent for 1 hour 9 minutes on 26 August, Workforce Planning on 4 August), and the Public API component is named once, for 2 hours 6 minutes of slowness on 3 August, which we scored between minor only and one major outage (12). Per-endpoint limits are published, such as 50 a minute on people search and 10 on employee updates (15). 429 carries Retry-After and X-RateLimit headers and the docs ask for backoff on 429 and 500, but no idempotency keys were found for writes (10). No SLA found, and the subscription terms disclaim uninterrupted service (0). The API is generally available, while the OAuth MCP server is in gradual rollout (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 4. Security & auth, 68 out of 100, up to 5.6 more on the total

Why it scored 68: Service users are separate, revocable credentials with no permissions by default and rights granted per feature, per field and per employee population. They travel as HTTP Basic. OAuth with 28 scopes, 5-minute access tokens and a customer-chosen audience exists for approved partners, and the MCP server uses OAuth as the signed-in employee (25). Field-level View and Edit make read-only service users possible. No confirmation step for terminate or delete calls was found (14). The API returns text written by employees, and no injection guidance was found. The MCP documentation was unreadable (2). Bob has an API audit log linked from the docs, whose article we could not open (10). SOC 2 Type II, ISO 27001:2022, ISO 27018:2019, third-party penetration tests and a Bugcrowd bounty. security.txt could not be checked (17).

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. Schema & documentation, 78 out of 100, up to 3.6 more on the total

Why it scored 78: Each endpoint and webhook page embeds an OpenAPI 3.1.1 definition in its Markdown version. No single downloadable spec was found (20). llms.txt and a .md twin of every page (10). Endpoint descriptions state permissions, supported user types and a list of critical behaviours such as no pagination and silent omission, though not when to choose another endpoint (15). Inputs are typed with enums (humanReadable APPEND or REPLACE) and limits (400 fields), but field IDs are free strings resolved through metadata endpoints (10). Request examples and documented 400, 403, 429 and default responses, with error bodies that differ by module (11). A /v1 path and a dated changelog with an RSS feed, with no per-entry dates in the Markdown twins (12).

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.

## 6. Maintenance & community, 59 out of 100, up to 3.6 more on the total

Why it scored 59: The newest developer changelog entry is dated 7 October 2026 (30). Ten entries between 21 July and 7 October, including new Employee Tables endpoints on 17 August and a Skills API on 5 August (20). A dated changelog with RSS and a support desk. No public issue tracker or developer forum was found (9). No official SDKs found, and the MCP registry could not be reached to check for an entry (0). No public packages or CI to assess (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 7. Transparency & trust, 71 out of 100, up to 2.5 more on the total

Made of editorial 58, provenance 83.

Why it scored 71: Closed service with public API Terms of Use and customer subscription terms revised January 2026 (15). The subscription terms promise deletion of personal data within 30 days of termination and let HiBob use anonymised, aggregated customer data freely. The privacy policy excludes data processed for customers, and the DPA sits behind a DocuSign link we could not read (15). Past deprecations carried dates and four to seven months' notice (2023 and 2024), but the deprecations page was last updated in May 2025 and the API terms promise notice only where commercially reasonable (10). The sub-processor list (November 2025) names providers, purposes and locations, including OpenAI and Azure OpenAI for optional functions, and hosting is stated as AWS Ireland with Frankfurt for recovery (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 4 of the 7 things a reader expects, and has 2 clauses that cost points (3.4 of 10)
- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)
- security.txt: could not be fetched (0 of 10)

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: the help centre article on the MCP server (server URL, tool list, annotations, confirmation of writes), which returned 403
- unchecked: www.hibob.com/.well-known/security.txt, which returned a Cloudflare block page
- unchecked: the DPA text, which sits behind a DocuSign link
- unchecked: the official MCP registry, which did not answer from our shell
- unchecked: the help centre article on API audit logs and on rotating service user tokens
- Whether a time off request can be approved or declined through the public API. No such endpoint is in the reference index, and the MCP tool list was unreadable
- Whether API access or the MCP server costs extra on top of the core subscription
- Whether an SLA exists in order forms
- apidocs.hibob.com served a bot check to curl after about 25 requests. Later pages were read through WebFetch

## Weaknesses

- No public price, free tier or trial. The sandbox is a purchased module
- People search has no pagination and returns every matching employee in one response
- Fields without permission or with invalid IDs are dropped from a 200 response with no warning
- No idempotency keys and no official SDK found in the reviewed documentation
- MCP setup and tool documentation sit in the help centre, which returned 403 to our reader

## 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.

- Send `Authorization: Basic base64(SERVICE-USER-ID:TOKEN)` to https://api.hibob.com/v1. Ask the Bob admin to put the service user in a permission group first, because it starts with none
- Request only the fields needed in `fields` on POST /people/search (maximum 400). The call returns all matching employees at once, so batch by `root.id` in large companies
- Compare returned fields with requested ones. Missing permission or a wrong field ID yields 200 with the field omitted
- Stop on 401 or 403. More than 50 in 10 seconds blocks the IP for 5 minutes
- Back off on 429 using Retry-After. Writes such as update, create and terminate employee allow 10 calls a minute, and no idempotency key exists, so check state before retrying a write

## 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.