Bullhorn

# Fix list: Bullhorn

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/bullhorn, the October 2026 research run, assessed 8 October 2026. Grade D, 46.7 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 Bullhorn: 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. Security & auth, 31 out of 100, up to 12.1 more on the total

Why it scored 31: OAuth 2.0 authorisation code grant with a client ID and secret issued by Bullhorn, ten-minute access tokens and rotating refresh tokens, exchanged for a `BhRestToken` session. No scopes were found, so a session carries all the rights of the API user (20). The guide shows the username, password, client secret and access token as URL query parameters, and `BhRestToken` may be sent in the query string, which takes 10 off (10). Access follows the API user's entitlements per entity and action, readable at `/entitlements`. No read-only mode or confirmation step for deletes or `/massUpdate` was found (8). Candidate CVs and notes are untrusted text, and no guidance on handling them was found. The fair use policy requires written permission before any AI tool is connected, which is a contract term and not a mitigation (0). `EntityEditHistory` records field changes and events carry the user and transaction. No per-call API log for the operator was found (5). The GDPR commitment statement reports annual SOC 1 Type 2 and SOC 2 Type 2 audits. No security.txt, disclosure policy or bug bounty was found, and www.bullhorn.com/security describes only the Salesforce-based Talent Rover product (8).

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.

## 2. Payments & pricing, 5 out of 100, up to 11.9 more on the total

Why it scored 5: Read with the hosted rubric. No x402, MPP or L402 (0). The pricing page gives Bullhorn Starter at $99 and Bullhorn Core at $165 per user per month and quotes the other plans. The partner route to the API is an annual platform fee with overage charges, neither published (5). The small agency pricing page says Bullhorn has no free trial at present, and the partner FAQ says free access is given to nobody (0). Customers get OAuth keys through a support ticket and partners through a signed contract, an invoice and a security assessment (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. Reliability, 53 out of 100, up to 9.4 more on the total

Why it scored 53: Read with the hosted rubric, for the REST API. Statuspage at status.bullhorn.com with 96 components by product and cluster and incident history, but no REST API component (20). From 12 July to 7 October 2026 the page lists 50 incidents across all products, 37 minor, seven major and two critical. On Bullhorn ATS/CRM clusters the majors were performance issues on CLS43 for 2 hours 3 minutes on 26 August, on CLS40 for 22 minutes on 27 August, and a Greenshades integration fault from 1 to 2 September. We read that as one major outage of the core product on one cluster (10). No numeric rate limit is in the API reference. The partner FAQ gives an allowance of 200,000 calls a month, and the guide says login rates are strictly limited without a figure (5). The error table says to wait one second on 429 and retry until successful. No idempotency keys were found for writes (8). No SLA was found on the public site (0). The API is generally available at version 2.0 (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. Schema & documentation, 55 out of 100, up to 7.3 more on the total

Why it scored 55: No OpenAPI or other downloadable contract was found. `/meta/{entityType}` returns field names, types and lengths to an authenticated caller, and `@bullhorn/bullhorn-types` publishes TypeScript interfaces for the entities (8). The docs are one HTML page built from Markdown sources in a public GitHub repository, and www.bullhorn.com/llms.txt is a product index that links to the developer site. No Markdown is served for agents (4). Each operation states its purpose with parameter tables, and the entity guide describes 91 entities field by field (12). Parameters are marked required or optional and entity fields carry types and lengths, but `query` and `where` are free-form Lucene and JPQL strings (9). Every operation has a curl example and a sample response. Errors are a table of ten status codes with no bodies (9). Login takes a `version` parameter, and a change log lists releases from 2024.5 to 2026.6 (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.

## 5. Agent ergonomics, 64 out of 100, up to 5.9 more on the total

Why it scored 64: Every read must name its `fields`, with nested sub-fields and a count per to-many association, and `fields=*` is blocked outside `/meta` (22). `start` and `count` paging with a default of 20 and a maximum of 500 (300 on some calls), Lucene search, JPQL where clauses and `orderBy` (20). Errors are status codes with one-line meanings, and the reference shows no error body to branch on (8). No idempotency keys. The only retry guidance is one second on 429, and the event queue can re-read a batch by `lastRequestId` (4). Official Java and TypeScript libraries exist, but a first call takes four requests (data centre lookup, authorisation code, token, REST login) and the base URL differs per customer (10).

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

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

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

## 6. Maintenance & community, 56 out of 100, up to 3.9 more on the total

Why it scored 56: The REST API change log's newest entry is the June 2026 release (2026.6), over 90 days old. The docs repository had commits on 17 September, 1 October and 5 October 2026, and `sdk-rest` 3.0.0 was tagged on 29 September 2026. We scored between the two (20). Three `sdk-rest` tags in 90 days (2.6.3 on 12 August, 2.7.0 and 3.0.0 on 29 September) and no dated API change log entry in that time (10). Closed service with a public change log, a developer ticket form and forums named in the guide. We could not read the help centre (6). Current official libraries in Java and TypeScript (15). `@bullhorn/taurus` 3.9.8 dates from 22 April 2026 with 51 weekly downloads, and `sdk-rest` moved to Spring 7 and Java 17 in September (5).

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

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

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

## 7. Transparency & trust, 67 out of 100, up to 2.9 more on the total

Made of editorial 51, provenance 82.

Why it scored 67: Closed service. The Master Subscription Agreement is not published, and the legal terms page links only to product supplements. The API Fair Use Policy is public and lets Bullhorn suspend access or discontinue the APIs at any time (10). The privacy policy of 20 May 2026 covers the service platforms, lists machine learning and product enhancement among its purposes, and keeps data while an account is active with no fixed periods. No public DPA was found beyond the standard contractual clauses pages (18). No deprecation policy was found. The reference notes that `fields=*` is deprecated and will be blocked, without a date (3). The sub-processor list, version 25 August 2026, gives each provider's purpose, location and covered services, with a form to subscribe to changes (20).

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 2 of the 7 things a reader expects, and has 2 clauses that cost points (1.7 of 10)
- security.txt: not found (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.

- The API Fair Use Policy of 17 December 2025 forbids connecting third-party AI or LLM tools, or MCP, without Bullhorn's explicit written permission. How that permission is requested and how often it is granted was not found
- unchecked: the API usage limits the fair use policy refers to. The reference gives no number, and help.bullhorn.com did not answer our requests
- unchecked: the Master Subscription Agreement, the API Access Agreement, any SLA and any DPA. None is published on the pages we read
- unchecked: the annual platform fee and overage price for partner API access, and whether customer API keys cost extra on each plan
- unchecked: which Bullhorn products hold SOC 2 Type 2 or ISO 27001. The GDPR commitment statement says many do without naming them, and trust.bullhorn.com did not answer
- unchecked: the 90-day window before 12 July 2026. The status feed returns the 50 most recent incidents, which reach back to that date
- unchecked: security.txt. The path redirects to the home page, which we recorded as none
- Not found in the reviewed documentation: an OpenAPI file, OAuth scopes, idempotency keys, error response bodies, a deprecation policy and an official MCP server

## Weaknesses

- The API Fair Use Policy of 17 December 2025 bars connecting third-party AI or LLM tools, or MCP, without Bullhorn's explicit written permission
- Credentials come from a support ticket for customers or a paid annual partner contract with a security assessment. No self-serve key, trial or free sandbox
- The documented flow puts the username, password, client secret, access token and `BhRestToken` in URL query strings
- No OpenAPI file, no OAuth scopes, no idempotency keys and no numeric rate limit were found in the reviewed documentation
- The status page listed 50 incidents from 12 July to 7 October 2026, seven marked major and two critical, with no API component

## 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.

- Confirm the customer holds Bullhorn's written permission for AI or LLM access before any call. The API Fair Use Policy forbids it otherwise
- Call `/rest-services/loginInfo?username=` first to learn the data centre, then use the returned OAuth and REST URLs and follow any 307 redirect
- Log in once and reuse `BhRestToken` until a call returns 401, then use the refresh token. Bullhorn limits login rates and may block frequent logins
- Send `BhRestToken` as a header, never in the URL, and always pass `fields`. `fields=*` is blocked outside `/meta`
- On 429 wait one second and retry. Writes have no idempotency key, so read the record back before repeating a PUT

## 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.