# Readiness scorer — can an agent sign up to your SaaS?

> Probe a signup URL you own and score whether an AI agent can sign up unaided.

- name: io.github.agentwares/readiness-scorer
- version: 0.1.2
- connection class: R0 (autonomous)
- trust: 90/100 flags: duplicate-repo
- quality: 100/100 (strong)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.agentwares%2Freadiness-scorer
- tools: https://api.protogrid.dev/v1/servers/io.github.agentwares%2Freadiness-scorer/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "readiness-scorer": {
      "type": "http",
      "url": "https://agentwares-readiness.vercel.app/api/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 100
- freshness: 85
- provenance: 90
- drivers: +repository +namespace-matches-repo +website +reachable +uptime-100% ~updated-39d-ago -duplicate-repo(7 names)
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 100/100 (strong)

### protocol: 100 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- pass `protocol.list_ttl`: tool list is cacheable (ttlMs 300000)

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 100 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 3 of 3 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 507 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: 3 tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 3, about 507 tokens to load them all
- tool set hash: c8e7bdea0eb328e9f023c4fbda3bdd0d
- tools last changed: 2026-10-09T07:40:35.182Z
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.agentwares/readiness-scorer.svg)](https://protogrid.dev/servers/io.github.agentwares/readiness-scorer)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-09 `readiness_get_scan` changed (annotations)
- 2026-10-09 `readiness_request_scan` changed (annotations)
- 2026-10-09 `readiness_get_verification_token` changed (annotations)

Full history: https://api.protogrid.dev/v1/servers/io.github.agentwares%2Freadiness-scorer/changes

## Remotes

- https://agentwares-readiness.vercel.app/api/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (3)

- `readiness_get_scan`: 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.
- `readiness_get_verification_token`: 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, <meta> tag, /.well-known file). The token is deterministic for the host, so calling this again returns the same value. Nothing is stored.
- `readiness_request_scan`: 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.