# Fix list: framefields From Anchor Terminal's listing at https://www.anchorterminal.com/tools/framefields, the October 2026 research run, assessed 20 October 2026. Grade C, 61.3 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 framefields: 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, 45 out of 100, up to 9.6 more on the total Why it scored 45: Read with the framework lines. The library takes no credentials. The worker uses one bearer key for all jobs plus a git token (15). The worker refuses repositories outside a required `FF_REPO_ALLOWLIST`, removes secrets from the environment before project code runs, shares node_modules read-only, and its README warns that allowed repositories' code runs in the worker's process. The getting-started page has the owner deny five of six dependency install scripts (12). It renders the owner's assets and returns no third-party text, apart from preview notes a person writes for the agent (10). Worker results report the rendered commit SHA, timings and GPU; no audit log (5). No SECURITY.md, no security.txt (404) and no npm provenance attestations on 2.3.2. Vision models are fetched from pinned Hugging Face revisions with a SHA-256 check (3). 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. Reliability, 61 out of 100, up to 7.8 more on the total Why it scored 61: Read with the local-software lines, because the library and the renderer worker run on the owner's machines. Official npm package `framefields` with Node.js 22 or later stated in the README and getting-started page, though not in `engines` (20). The one public CI workflow builds, typechecks and tests the renderer worker on every push to main and passed on the last ten runs (14 to 19 October 2026). GPU tests are skipped there, and the library's own Vitest suite isn't run in CI (10). Seven bug issues filed on 13 October 2026 by an outside reporter were all closed with fixes within a day; one outside pull request is open. A young project with a small sample (18). Changesets changelog with major, minor and patch entries, but 2.2.0, a minor release, removed exports from `@framefields/client-utils` and changed the arguments of `mergeSubjectMask` (8). Version 2.3.2, but the site and README declare it beta, so 5 of 15 (5). 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. ## 3. Transparency & trust, 39 out of 100, up to 5.3 more on the total Made of editorial 51, provenance 27. Why it scored 39: Apache-2.0, an OSI licence, in LICENCE and the npm metadata (30). No privacy policy, but nothing is hosted. The docs say vision runs on the device; the source shows model downloads from Hugging Face and a preview page that loads Google Fonts and onnxruntime-web from jsDelivr, which the reviewed docs don't state (8). No deprecation policy; the changelog names minimum dependency versions and removed exports, without notice periods (3). No telemetry found in the source and none mentioned in the docs; no statement either way (10). 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): - Legal entity named: not found (0 of 20) - Domain age: framefields.dev, no registry record we could read (0 of 15) - Status page: not found (0 of 10) - security.txt: not found (0 of 10) ## 4. Payments & pricing, 60 out of 100, up to 5 more on the total Why it scored 60: Read with the self-hosted rule. Free Apache-2.0 software with nothing to buy from Gatewai and no payment protocol (0). No price to publish (20). No card or trial needed (20). An agent can install and run the package with no sign-up; the worker runs on a host the owner already has (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. ## 5. Agent ergonomics, 74 out of 100, up to 4.2 more on the total Why it scored 74: A library with no tool definitions to load. The four skills are 88 KB of Markdown and load only when a task matches (16). Output can be sized to the check: one frame, a contact-sheet grid with a step, a frame range, a quality level, and on the worker an LFS include list (16). `comp.lint()` returns `{ severity, code, message, layerId }`, `renderVideo({ qa })` returns measured issues with times, and worker jobs fail with `{ code, message, phase, retryable, hint }` (18). Frames are a function of the frame number; worker jobs take an `idempotencyKey` and a `source.commit` pin, and renders accept an AbortSignal (16). Setup needs Node.js 22, an ES module project and an approved skia-canvas install script; TypeScript only (8). 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. Schema & documentation, 77 out of 100, up to 3.7 more on the total Why it scored 77: Read as a library. No OpenAPI or other formal contract; the package ships TypeScript declarations for every export and validates with zod, counted as 15 of 25 as for other SDKs (15). llms.txt at framefields.dev and a docs index at /docs/llms.txt with a Markdown version of every page and full and compact text files (10). Authoring rules, a workflow page and four skills say what to do and in what order, cheapest check first (15). Options are typed with enums, such as `quality` (very_low, low, high, very_high) and `codec` (h264, vp9, vp8), and job fields have defaults (12). Examples on every guide page; lint and QA codes, troubleshooting and worker error codes are each catalogued with a fix (14). Semantic versions with a changesets CHANGELOG.md in the repository; no GitHub releases and no changelog page on the site (11). 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, 89 out of 100, up to 1 more on the total Why it scored 89: Version 2.3.2 published on 19 October 2026, the day before the check (30). 22 npm versions since 5 October 2026 (20). Outside bug reports of 13 October 2026 were fixed and closed within a day, and an outside pull request of 13 October has three comments and is still open (18). It is Gatewai's own package, current with the skills at the same version (15). Dependencies are current (zod 4, mediabunny 1.61.6 as a stated minimum), but CI covers only the worker and the GPU backend depends on `webgpu` 0.4, a pre-1.0 package (6). 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. - 20 October 2026. The install page says `/plugin install framefields` comes from Anthropic's official plugin directory, which Claude Code adds for you. The directory's public marketplace file (last commit 19 October 2026) lists 315 plugins and no framefields entry, so the command as documented would not find the plugin. It may be a submission still pending; the deduction lapses if the entry appears. Sources: https://framefields.dev/docs/agents/install/ and https://raw.githubusercontent.com/anthropics/claude-plugins-official/main/.claude-plugin/marketplace.json (-2) ## 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 npm package page at www.npmjs.com, which answered with a Cloudflare bot check (HTTP 403); the registry document was read instead. - The Product Hunt launch of 19 October 2026 was not read first-hand; firstReleased is the first npm version, 5 October 2026, two weeks before it. - Whether the Claude Code official-directory entry is pending or was never submitted. - No legal entity for Gatewai was found; gatewai.studio is drawn by script and has no readable text. - Installation, render output, speed and the vendor's memory and determinism figures were not tested, by rule. ## Weaknesses - First published on 5 October 2026 and declared beta - No security policy, security.txt or npm provenance attestations found - Public CI runs the worker's tests, not the library's Vitest suite - Needs Node.js 22, an approved skia-canvas install script and ideally a GPU ## 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. - Add the plugin with `/plugin marketplace add gatewai-dev/framefields` first; the official-directory install was not found on 20 October 2026. - Approve only the skia-canvas install script, as the getting-started page shows, or imports fail with a missing `skia.node`. - Check with `comp.lint()` and `renderFrameGrid` before a full `renderVideo` encode. - Call `process.exit()` after rendering; native GPU threads keep the script alive. - Preload vision models into `FRAMEFIELDS_MODELS_DIR` for offline or CI renders. ## What the review panel asked for - keep removals to major versions - dated deprecation notices before removals - Fix the plugin install line - Mark breaking changes as major - library tests in CI - a stated deprecation policy - isolate job code from worker secrets - a published security policy ## 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.