HoneyHive by HoneyHive

# Fix list: HoneyHive

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/honeyhive, the October 2026 research run, assessed 1 October 2026. Grade C, 55.9 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 HoneyHive: 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, 44 out of 100, up to 11.2 more on the total

Why it scored 44: Better Stack page at status.honeyhive.ai with a single component, Backend Platform, and nothing separate for the data plane, control plane or ingestion (12 of 20). The page showed 80.992 per cent uptime for that component and the banner "Some services are down" when we loaded it at 00:59 UTC on 2 October, with no incident write-ups for the last 90 days. The only write-up we found was a SOC 2 incident-response drill from March 2025. We can't tell whether the monitor watches a retired host, but the vendor's own page shows the figure, so 0. The pricing page publishes 1,000 requests a minute on Free and custom limits on Enterprise, and the changelog gives a 7 MiB ingestion body limit (12). The OpenAPI specs declare no 429 and we found no `Retry-After` or backoff guidance. The Python SDK has a configurable retry budget (5). Enterprise includes an uptime SLA with service credits, terms not published (5). API, SDKs and CLI are 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, 20 out of 100, up to 10 more on the total

Why it scored 20: No x402 or other machine payment (0). The only paid plan is Enterprise at a custom price, so there's no public per-unit or plan price (0). The free Developer plan needs no card (20). A person signs up in a browser to get a key (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. Security & auth, 61 out of 100, up to 6.8 more on the total

Why it scored 61: Four key types. Project keys (`hh_`), read-only project keys (`hh_ro_`) since 11 June, fine-grained control plane keys (`hh_fgcp_`) with chosen permissions and a required expiry of at most 365 days, and ingestion-only keys (`hh_ingst_`) bound to one project since 17 September. Revocation takes effect within two minutes (30). Read-only and ingestion-only keys give least privilege, and deleting a workspace with active children fails with 409 unless `--dangerously-delete-child-scopes` is passed (18). Traces hold whatever the application logged, and we found no prompt-injection guidance for agents reading them (3). No audit log found (0). The pricing page claims SOC 2 Type II, GDPR and HIPAA. The Drata trust centre at trust.honeyhive.ai rendered nothing readable to us, there's no security.txt, and the CLI changelog lists "Security fixes" in 1.6.0 without detail (10).

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.

## 4. Agent ergonomics, 63 out of 100, up to 6 more on the total

Why it scored 63: No MCP server for platform data. The docs MCP only searches documentation. Event search defaults to 1,000 rows with no field projection, so an agent has to set `limit` itself (10). Filters, date ranges, `limit` and `page` on search (18). Since 24 September a bad key, a revoked key and a missing permission all return the same 404 as a missing resource, which leaves an agent guessing (10). `POST /session/start` is idempotent on `session_id`. We found no other idempotency keys (10). Python SDK, TypeScript SDKs and a CLI generated one to one from the specs with agent skills. Few required parameters (15).

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.

## 5. Transparency & trust, 68 out of 100, up to 2.8 more on the total

Made of editorial 54, provenance 82.

Why it scored 68: Closed platform. The Python SDK, CLI and OpenAPI specs are MIT (18). The privacy policy (11 January 2025) covers platform data, defers retention to the pricing page (30 days on Free), names AWS and Stripe and uses standard contractual clauses for EU transfers. Self-hosted data-flow docs exist and a DPA is custom on Enterprise (14). 22 operations marked deprecated in the spec and changelogs with Compatibility and Deprecations sections that say when flags go. Against that, the v2.0.0 spec of 8 May removed `GET /events` and nine other operations without deprecation, per its own oasdiff changelog (14). AWS hosting and US transfer stated, no full subprocessor list (8).

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):

- Domain age: honeyhive.ai, registered 2022-07-28 (4 years) (7 of 15)
- security.txt: not found (0 of 10)

## 6. Schema & documentation, 86 out of 100, up to 2.3 more on the total

Why it scored 86: Two public OpenAPI 3.1 specs in an MIT repository, data plane with 45 paths and 70 operations and control plane with 8 paths and 15 operations (25). llms.txt with every page as Markdown (10). Every operation has a description, deprecated operations are flagged (22 of them) and `POST /v1/events/search` is labelled the primary way to read events (16). Typed request bodies with bounds such as `limit` 1 to 1,000 and `additionalProperties: false`, though event inputs, outputs and metadata are free-form objects (11). Only 9 operations carry examples, no 429 is declared, and since 24 September permission and key failures all return 404 (9). Spec changelog generated with oasdiff, plus dated product, client and self-hosting changelogs (15).

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.

## 7. Maintenance & community, 77 out of 100, up to 2 more on the total

Why it scored 77: Python SDK 1.6.1 on 2026-09-29 (30). Python SDK 1.5.0, 1.5.1, 1.6.0 and 1.6.1 and CLI 1.5.0, 1.5.1, 1.6.0 and 1.7.0 since 20 July (20). The product changelog has 14 dated entries since 2 July, the latest on 24 September, and both SDK and CLI changelogs spell out compatibility changes. We didn't load issue reply times (12). The Python SDK is current. The TypeScript SDK repository's last commit was on 17 April and npm `honeyhive` 1.0.45 dates from 9 July (10). The Python SDK repository runs only docs, publish and PR-title workflows, with no test workflow on pull requests that we could see (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.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 2026-09-22 to 2026-09-24. The API stopped returning 401 for invalid, revoked or expired keys and 403 for permission failures, and now answers 404 for all of them, on every client version. It was announced in the CLI 1.7.0 changelog on 22 September and the product changelog on 24 September, with no deprecation window. -3 (https://github.com/honeyhiveai/honeyhive-cli/blob/main/CHANGELOG.md)

## 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.

- What the status page's Backend Platform monitor watches, and whether its 80.992 per cent figure reflects the current v2 data plane
- unchecked: the Drata trust centre, so certification dates, subprocessors and any disclosure programme are unverified
- Whether the v1 API host still serves the operations the v2.0.0 spec removed on 8 May 2026
- Whether any audit log exists for API key use. We found none in the docs

## Weaknesses

- Status page showed 80.992 per cent uptime for its single component and "Some services are down" on 2 October, with no incident history
- No MCP server for trace data, only a docs search MCP
- No self-serve paid tier, only Free and a sales-led Enterprise plan
- Since 24 September 2026 invalid keys and permission failures return 404, with no deprecation window
- Event search returns up to 1,000 rows by default with no field selection

## 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.

- Give the agent a read-only `hh_ro_` key for analysis and a separate `hh_ingst_` key for sending traces
- Treat a 404 as possibly a bad, revoked or expired key before assuming the record is missing
- Set `limit` on `POST /v1/events/search`. The default is 1,000 rows
- Use the CLI (`brew tap honeyhiveai/tap && brew install honeyhive`) from a coding agent. It maps one command to each API endpoint
- Control plane commands need an `hh_fgcp_` key and the control plane host api.cp.us.honeyhive.ai

## What the review panel asked for

- notice before behaviour changes
- restore 401 and 403
- declare 429 responses

## 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.