Boski

# Fix list: Boski

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/boski, the October 2026 research run, assessed 14 October 2026. Grade F, 26.2 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 Boski: 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, 25 out of 100, up to 15 more on the total

Why it scored 25: Read with the hosted lines. No status page found. status.boski.com answers with a hosting not-found page (0). No readable incident history (5). The terms set one plan with a weekly allowance of 5,000 credits, the Marmot reading for plan limits with numbers and no per-call figures (10 of 15). No 429 or retry guidance for the plugin (0). No SLA (0). Launched to everyone on 11 September 2026 and not labelled beta, though the price is called experimental (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. Agent ergonomics, 13 out of 100, up to 14.1 more on the total

Why it scored 13: Graded on the plugin. The number and size of its tools are unchecked, so the context-cost line earns nothing (0 of 25). No paging or size controls documented (0). One documented failure, an unfinished interview, with its fix (2 of 20). Access is read-only, so a repeated call changes nothing, with annotations unread (8 of 20). The plugin signs in with the Boski account and documents no parameters, and there's no SDK (3 of 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.

## 3. Schema & documentation, 15 out of 100, up to 13.8 more on the total

Why it scored 15: Graded on the Boski plugin for ChatGPT and Codex, the one route an outside agent has into Boski. boski.com/.well-known/oauth-protected-resource names an MCP resource at www.boski.com/api/mcp with the scope personal_context:read. robots.txt disallows /api for every agent, so the server wasn't requested and its tool definitions are unchecked, the Pally reading for schemas we couldn't read (0 of 25). llms.txt and llms-full.txt (10). One support paragraph says the plugin reads saved preferences and constraints to tailor advice and plans and can't change them, with no tool names or guidance on when not to use it, the Khoj dossier's 3 for one-line descriptions (3 of 20). Typed inputs unchecked (0 of 15). One troubleshooting case (finish the onboarding interview, then retry) and reconnect steps, with no examples or error list (2 of 15). No version or changelog for the plugin (0 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.

## 4. Payments & pricing, 10 out of 100, up to 11.3 more on the total

Why it scored 10: Read with the hosted rule. No x402, MPP or L402 (0). The launch post of 11 September 2026 sets experimental pricing of $5 a week, and the terms describe one plan with a weekly billing period and 5,000 credits, priced at checkout, plan-only (10). No free plan or trial found (0). A person signs up through Boski's web checkout (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.

## 5. Security & auth, 45 out of 100, up to 9.6 more on the total

Why it scored 45: Graded on the plugin. OAuth through boski.com/api/auth with a single read scope, personal_context:read, and the protected-resource metadata lists DPoP signing algorithms for sender-constrained tokens. Disconnecting is done in the host app, and client registration and token revocation weren't readable. Pally's single-scope OAuth got 22, and this is the same kind (22 of 30). The support page says access is read-only and can't save changes, so the plugin has no write actions (15 of 20). It returns the owner's saved preferences, which may include information Boski matched from public professional or social sources when the profile was created, per the privacy policy, with no injection guidance (8 of 15). No per-call log of plugin access found (0 of 15). boski.com/.well-known/security.txt returns 404, and no security contact, disclosure policy, bug bounty, certification or advisories were found. Support is contact@boski.com (0 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.

## 6. Maintenance & community, 23 out of 100, up to 6.7 more on the total

Why it scored 23: Read for a closed service. No changelog. The newest dated product post is the launch to everyone on 11 September 2026, 33 days before this check (20). Two dated product posts in the last 90 days, 10 August (iMessage) and 11 September (0). A blog and a support page with contact@boski.com, untested (3 of 15). No MCP registry entry or SDK for the plugin (0). No package or CI to judge (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, 63 out of 100, up to 3.2 more on the total

Made of editorial 50, provenance 76.

Why it scored 63: Closed service under terms effective 20 August 2026 naming Boski, Inc., Delaware (15). The privacy policy of 4 September 2026 says Boski doesn't train its own or third-party foundation models on user content or let AI providers do so, may use it for narrow internal systems such as a model-selection router, and may match public information about a user at profile creation. Consistent statements without retention periods (18 of 30). No deprecation policy, and the launch price is called experimental (2 of 20). Providers named include Vercel, Neon, Vercel AI Gateway, Anthropic, OpenAI, Google, Resend, AgentMail, Meta, Twilio, Apple, Linq, X and Stripe, with no regions (15 of 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):

- Status page: not found (0 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.

- unchecked: the plugin's MCP server at www.boski.com/api/mcp and its authorization server under /api/auth, which robots.txt disallows, including its tools, client registration, PKCE and revocation.
- Whether agents other than ChatGPT and Codex can connect.
- Retention periods for user content.

## Weaknesses

- No free plan or trial found.
- The plugin's endpoint isn't documented, and its tool definitions weren't readable.
- No status page, security.txt or security contact found.
- Boski may search public sources to match information about a new user, per the privacy policy.

## 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.

- Use the Boski plugin only to read the owner's saved preferences and constraints. It can't save changes.
- Ask the owner to finish Boski's onboarding interview if the plugin says it's incomplete, then retry.

## 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.