FlightClaw by FlightClaw

# Fix list: FlightClaw

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/flightclaw, the October 2026 research run, assessed 1 October 2026. Grade E, 45.5 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 FlightClaw: 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, 30 out of 100, up to 14 more on the total

Why it scored 30: No status page found on flightclaw.com, in the docs or in llms.txt (0). No incident history to read (5). Limits published with numbers, 100 flight searches per user per UTC day and 120 requests a minute on the hosted MCP, 20 searches and 10 checkouts a day per IP on the website (15). No 429 behaviour, Retry-After or retry guidance found, and nothing on retrying `create_checkout` (0). No SLA (0). Nothing on the site, docs, llms.txt or registry entry calls the hosted server beta, and it's listed as 1.0.0 in the official registry (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, 50 out of 100, up to 8.8 more on the total

Why it scored 50: OAuth 2.1 with PKCE (S256), dynamic client registration and a revocation endpoint, per the server's authorisation metadata. One scope, `flights`, covers search, booking and the profile store (25 of 30). No read-only mode, but payment needs a person to pay on the checkout link or approve an exact-amount Link spend, so the agent can't spend on its own (12 of 20). Results are airline offer data, with some user-written content in trip notes and profiles. No injection guidance (8 of 15). `list_orders` and `get_checkout_status` show what happened; the privacy policy says request logs (IP, time, tool name) are kept for security, with no operator view (5 of 15). No security.txt (404 per the 30 September check), disclosure policy, bug bounty or certification found (0).

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, 47 out of 100, up to 8.6 more on the total

Why it scored 47: The README lists 36 hosted tools in five groups, llms.txt lists 18 and the docs page fewer. We couldn't call tools/list without signing in, so we score between the 11 to 30 and over-30 bands (10 of 25). Search takes cabin, stops, airlines, price, duration, time and layover filters and a result count (12 of 20). No documented error codes; the local server returns plain-text messages such as "not configured" (5 of 20). The local server marks every tool with readOnlyHint, destructiveHint or idempotentHint, with both cancel tools destructive. The hosted annotations are unchecked (12 of 20). Three required inputs for a search and sensible defaults, no SDK (MCP only) (8 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.

## 4. Schema & documentation, 50 out of 100, up to 8.1 more on the total

Why it scored 50: The hosted server's tool definitions sit behind OAuth and weren't readable. The open-source local server builds typed JSON Schema from Python type hints on every tool (15 of 25). llms.txt at flightclaw.com/llms.txt with the flow, limits, pricing and FAQ (10). Local docstrings say what each tool does and the llms.txt spells out search, get_offer, create_checkout, pay on the link, but there's little on when not to call a tool (12 of 20). Cabin, stops and sort are plain strings with the allowed values in the docstring rather than enums (8 of 15). Example prompts in the README, no documented error responses (5 of 15). Version numbers in the manifests (1.0.0 in the registry, 1.1.0 in server.json) but no changelog (0).

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. Payments & pricing, 50 out of 100, up to 6.3 more on the total

Why it scored 50: No x402, MPP or L402 (0). The fee is published without login, 5 per cent plus $3 per booking with a $12 minimum, $6 per change, and the API is free (20). Search and the MCP cost nothing and no card is needed to start (20). Sign-in needs a person to read a 6-digit email code. Browser agents can use the WebMCP tools on flightclaw.com with no sign-in, 20 searches and 10 checkouts a day, which is a keyless route of sorts (10 of 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.

## 6. Transparency & trust, 32 out of 100, up to 6 more on the total

Made of editorial 28, provenance 35.

Why it scored 32: The local server is MIT. The hosted service runs under a short terms page that names no company, registered address or governing law, so it isn't clear who the contract is with (10 of 30). The privacy policy lists what's collected (email, travellers, preferences, loyalty balances, request logs) and says card numbers go to Duffel's form and are never stored. No controller, no DPA, and retention is "until you ask us to delete it" (10 of 30). No deprecation policy or dated notices (0). Duffel and the AI assistant providers are named as recipients; no data locations (8 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):

- Legal entity named: not found (0 of 20)
- Domain age: flightclaw.com, registered 2026-02-10 (under a year) (0 of 15)
- Status page: not found (0 of 10)
- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 68 out of 100, up to 2.8 more on the total

Why it scored 68: Last change 2026-09-26, the registry publication and the hosted-MCP README (30). Commits on 30 August, 2 and 26 September and a v1.0.0 tag on 2 September, but no changelog or release notes (10 of 20). Two outside pull requests (#1 and #4) were merged on 3 June. GitHub's issue pages are closed to our reader, so open issues are unchecked (10 of 25). In the official MCP registry as com.flightclaw/flightclaw, a domain-verified namespace (15). No CI workflow in the repository; the local server pins its dependencies (3 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.

- Who operates the service. The terms, privacy policy and site name no company; the repository licence names an individual developer
- The hosted tool list. The README says 36 tools, llms.txt lists 18 and the docs page fewer; we couldn't call tools/list without signing in
- Whether the hosted tools carry the same readOnlyHint and destructiveHint annotations as the local server
- We removed the beta tag because nothing we read calls the service beta, and added llms-txt for the file at flightclaw.com/llms.txt
- Open GitHub issues and reply times, since GitHub's issue pages are closed to our reader

## Weaknesses

- Terms and privacy policy name no legal entity, address or governing law, and Companies House has no FlightClaw company
- No status page, changelog, SLA or documented error responses
- One OAuth scope covers search, booking and the stored traveller, card-label and loyalty profile
- 100 searches per user per UTC day on the hosted server
- Tool count differs between the README (36), llms.txt (18) and the docs page

## 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.

- Call `get_offer` right before `create_checkout`; the total includes the booking fee and can move
- Poll `get_checkout_status` until `completed`, then `get_order` for the airline reference. A sent link isn't a booking
- Never ask for card numbers in chat. Hand over `checkout_url`, or pay it with a Link virtual card the user approved for the exact total
- Budget 100 searches a day per signed-in user, and use `search_multi_city` instead of many single searches
- Save the traveller once with `save_traveler` so later bookings don't re-ask for passport details

## What the review panel asked for

- Keyless route beyond browsers
- Name the contracting company
- Publish one tool list

## 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.