Honcho by Plastic Labs

# Fix list: Honcho

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/honcho, the October 2026 research run, assessed 1 October 2026. Grade B, 64.2 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 Honcho: 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: Rootly status page at status.honcho.dev with Honcho API and Web Dashboard components and history (20). The last 90 days hold scheduled maintenance on 4 August (about 2 hours 50 minutes, announced), slow provisioning of new instances on 17 August (about 7 hours, existing instances unaffected) and 36 minutes of migration downtime on 17 September. The page shows 99.96 per cent for the API over 60 days. Minor only (20). The May 2026 database outage (issue #753) falls outside the window. No published rate limits found (0). No 429 or retry guidance found (0). No SLA found (0). The v3 API is generally available (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. Security & auth, 48 out of 100, up to 9.1 more on the total

Why it scored 48: Keys can be minted through the API scoped to a workspace, a peer or a session, with an optional expires_at, and are revocable from the API Keys page. The MCP takes a Bearer key or OAuth (30). Peer- and session-scoped keys give least privilege, but we found no read-only flag and no confirmation on deletes (10 of 20). Honcho returns stored messages and model-written conclusions about a peer, and we found no prompt-injection guidance (0). No audit log or per-call log found (0). A SOC 2 Type I badge on the site. No security.txt, and no disclosure policy or bug bounty found (8 of 20).

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.

## 3. Agent ergonomics, 65 out of 100, up to 5.7 more on the total

Why it scored 65: The MCP tool count isn't published, so we can't score its context cost from the definitions. The API's context call returns a ready block for the next prompt (15 of 25). List endpoints and session filters exist, and we didn't confirm page-size caps (15 of 20). 422 responses name the failing field (12 of 20). POST /v3/workspaces gets or creates, so repeating it is safe, but we found no idempotency keys on message writes and no tool annotations (8 of 20). Python and TypeScript SDKs and few required fields (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.

## 4. Schema & documentation, 79 out of 100, up to 3.4 more on the total

Why it scored 79: OpenAPI specs for v1, v2 and v3 linked from llms.txt, at honcho.dev/docs/v3/openapi.json for the current version (25). llms.txt (10). Endpoint pages say what each call does, but the hosted MCP sends its tool list and instructions on connect instead of documenting them, so we couldn't read the descriptions (12 of 20). Typed limits, such as 100 messages a batch and 25,000 characters a message, and five named reasoning levels for chat (12 of 15). Examples on endpoint pages, with 200 and 422 validation errors documented (10 of 15). Versioned docs and a compatibility guide, but the changelog entries carry version numbers without dates (10 of 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.

## 5. Transparency & trust, 69 out of 100, up to 2.7 more on the total

Made of editorial 56, provenance 82.

Why it scored 69: The server is AGPL-3.0, the same code Honcho hosts, and the TypeScript SDK is Apache-2.0 (30). The terms keep data for 90 days after termination and then delete it, and a deleting-data page says what survives each kind of delete. We didn't re-read the privacy policy this run (18 of 30). Docs for v1, v2 and v3 sit side by side with a compatibility guide, but no dated notices or written policy (8 of 20). No subprocessor list or data-location statement found (0).

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: honcho.dev, registered 2022-01-26 (4 years) (7 of 15)
- security.txt: not found (0 of 10)

## 6. Payments & pricing, 80 out of 100, up to 2.5 more on the total

Why it scored 80: x402 and Tempo MPP on 16 endpoints at agentcash.honcho.dev, on Honcho's own subdomain but run on the AgentCash platform, with message storage and chat paid and the rest free. api.honcho.dev itself takes only keys (30 of 40). Ingestion at $2 per million tokens and chat at $0.001 to $0.50 a call are published (20). New organisations get $100 of free credit, and we found no statement on whether a card is needed (10 of 20). An agent can pay per call over x402 with no account (20).

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.

## 7. Maintenance & community, 73 out of 100, up to 2.4 more on the total

Why it scored 73: @honcho-ai/sdk 2.5.1 on npm on 2026-09-24 (30). The changelog isn't dated and the repo has no GitHub releases, so we couldn't count releases in the last 90 days (10 of 20, for one confirmed release and a moving changelog). A maintainer answered the May outage report (#753) and moved updates to Discord (10 of 25). The hosted MCP is in the official registry as io.github.plastic-labs/honcho, and the SDKs are current (15). The TypeScript SDK's only runtime dependency is zod. We didn't check CI (8 of 10).

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.

## 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 number of tools on the hosted MCP server isn't documented.
- We couldn't confirm whether the $100 free credit needs a card.
- The x402 storefront details, the 90-day retention clause and issue #753 come from the listing check on 2026-09-30. We didn't re-fetch them on 2026-10-01 because the fetch budget ran out.
- The listing carried a -3 deduction for the May 2026 outage. Outages aren't a negative-event category in the brief, and the reliability score covers the incident record, so we set it to 0.

## Weaknesses

- No published rate limits, 429 guidance or SLA
- Changelog entries carry no dates and the repo has no GitHub releases
- The MCP tool list is sent on connect, not documented
- No security.txt, audit log or prompt-injection guidance found
- AGPL-3.0 server licence

## 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.

- Get or create the workspace with POST /v3/workspaces before writing sessions and messages
- Batch up to 100 messages per request, each under 25,000 characters
- Use context retrieval for routine turns and save `chat` at high or max for questions that need it, since max costs 500 times minimal
- Mint a peer- or session-scoped key with an expiry for any agent that shouldn't see the whole workspace
- Over x402, batch messages rather than sending them singly, as each paid call has a $0.001 floor

## What the review panel asked for

- Document the MCP tools
- Add 429 guidance
- published MCP tool list
- a read-only key flag

## 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.