Probe a signup URL you own and score whether an AI agent can sign up unaided.
Scored across 3 tools
Each tool maps to a distinct, explicitly numbered step of a single workflow: token retrieval, scan request, and scan polling. There is no overlap or plausible way to confuse them, and descriptions reinforce the boundaries.
All three names use a consistent readiness_ prefix with snake_case verb_noun structure (get_scan, get_verification_token, request_scan). Only nit is that get_verification_token breaks the strict two-token verb_noun shape, but the convention is still clearly readable.
Three tools is lean, but for a narrowly scoped domain-ownership-and-scan workflow each tool is a necessary, non-redundant step. It is close to the 'thin' threshold but not underspecified for its purpose.
The full happy path is covered: obtain ownership token, submit/queue the scan, and retrieve results with score, rubric, and fix list. Minor gaps exist (no cancel/list-scans or bulk operations), but polling by id lets agents work around them.
Step 3. Poll a scan by id. While status is queued/running only metadata is returned; when done you get the 0-100 score, grade, per-item rubric points with evidence, the fix list, and what could not be observed.
Inputs: id (required).
Step 1 of a readiness scan. Returns the domain-ownership token for the URL's host plus the three ways to publish it (DNS TXT record, tag, /.well-known file). The token is deterministic for the host, so calling this again returns the same value. Nothing is stored.
Inputs: url (required).
Step 2. Verifies domain ownership (DNS TXT / meta tag / .well-known must carry the token from readiness_get_verification_token), then queues a Playwright probe that tries to sign up, accept the terms and mint an API key with no human. Returns the scan id and the public results URL. Fails with DOMAIN_NOT_VERIFIED (with instructions) or RATE_LIMITED (3 scans per host per day). Never solves CAPTCHAs.
Inputs: url (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Check a website in a real browser, or check if AI agents can use it: score, problems, report link.
Scan a public website for AI agent readiness: 60 checks, a score, and the evidence for each.
Scores any public website on how usable it is by AI agents, with per-check evidence.
Test whether AI agents complete SaaS pricing, signup or sales journeys; get evidence-backed reports.
Add one secure layer between your agents and this server.
