How to run the public scanners in one hour
An order of work for scoring one site on every public scanner, with the limits that bind each step.
In short
Start with cached lookups, then live scans. Run Ora, Cloudflare, is-agentic, agent-ready.dev and, for docs sites, Mintlify, then fill in the SiteSpeakAI form by hand. The limits that bind are Ora's 30 scans a day per IP and agent-ready.dev's 3 free scans per 30 days per IP. Record the time of every number.
Six scanners are public and only some can be scripted
This site covers six public scanners: Ora, Cloudflare's isitagentready.com, is-agentic.com, agent-ready.dev, SiteSpeakAI and mintlify.com/score. A person can try all six in a browser. They differ in what can be looked up, repeated or scripted, and in the limits that cap a run. The hour is a budget. The data gives no timings beyond about 30 seconds for a fresh Ora scan and "seconds" for an agent-ready.dev web scan, so this guide gives an order and limits, not minutes.
Work in this order, because cached results come first
| Step | Scanner | How | Limit that binds |
|---|---|---|---|
| 1 | Ora | GET /api/score/{domain} first. Only if it answers 404 DOMAIN_NOT_SCANNED, POST /api/scan; a fresh scan takes about 30 seconds. | 30 anonymous scans per IP per rolling 24 hours, 10 a minute, 6 that bypass the cache. Cached reads do not use the daily quota. |
| 2 | Cloudflare | The web tool, or POST /api/scan with a JSON body that names the url. Format "agent" returns one combined fix prompt. | No cap published. A site that answers the scanner with 403 gets no score. |
| 3 | is-agentic | npx is-agentic <domain> --json retrieves or starts a report. | Its llms.txt says automated bulk scanning is not permitted. Its public API only reads stored reports. |
| 4 | agent-ready.dev | The web scan, free with no sign-up, returns in seconds. | Anonymous POST scans: 3 per 30 days per IP, then an x402 payment. REST API and MCP need Pro. |
| 5 | Mintlify | Enter the docs URL at mintlify.com/score. Skip if the site has no docs site. | Scores documentation only, and only tracked docs sites. |
| 6 | SiteSpeakAI | Fill in the web form by hand and keep the share link. | No API, no lookup by domain and no public result list, so a result cannot be fetched later. |
The limits that bind decide how many sites fit
For one site only agent-ready.dev's allowance matters, and one scan costs one of three in 30 days. For several sites the order changes. A person with one IP address can score at most three sites a month on agent-ready.dev's anonymous scan, and at most 30 scans a day on Ora.
The 2026-11-04 Concordance shows why lookups first is the right habit. Of its 60 Ora cells, 58 came from the cached score and 2 from fresh scans. Its agent-ready.dev column is empty for all 60 domains, because the badge read "unscored" and the Concordance did not spend the 3 free scans.
- Head of web. Run steps 1, 2 and 4 and keep the dated results.
- Technical marketer. Add step 3 for the shareable report and step 6 once, by hand.
- Platform engineer. Script steps 1 and 2. Treat step 3 as Ora's result, not a separate one.
Record the time and source of every number
Cached scanners return the latest stored result, which can be older than your fetch. Record these for each cell:
- The score and the scanner's own band or grade.
- The time you fetched it.
- The scan date where the scanner exposes one. Ora and is-agentic report scanned_at, and Mintlify reports last_computed. agent-ready.dev's badge exposes no scan date.
- Which scan the number came from. Ora and is-agentic usually read the same underlying scan, so they are one measurement, not two. Cloudflare scans live at fetch time.
Ora's cached scores move within a day: Primitive showed 96 on the home page and 98 in /api/stats on 2026-11-04. Two readings an hour apart can differ without the site changing.
What this guide cannot tell you
- How long each step takes on your site. Only two durations are published.
- Whether a second run gives the same number. On 2026-11-05 a repeat run of three domains returned the same score on the four scanners that offer a lookup.
- Whether these limits will hold. All were read on 2026-11-04 and vendors change them.
- How to compare the six numbers. They sit on different scales; see why scanners disagree and the method page.
Sources
- Ora, llms-full.txt. https://ora.ai/llms-full.txt, read 2026-11-04.
- Ora, OpenAPI spec. https://ora.ai/api/openapi.json, read 2026-11-04.
- Ora, API stats. https://ora.ai/api/stats, read 2026-11-04.
- Cloudflare, isitagentready llms.txt. https://isitagentready.com/llms.txt, read 2026-11-04.
- is-agentic.com. https://is-agentic.com/, read 2026-11-04.
- agent-ready.dev, comparison with alternatives. https://agent-ready.dev/agent-ready-vs-alternatives, read 2026-11-04.
- Mintlify, Agent Score. https://www.mintlify.com/score, read 2026-11-04.
- SiteSpeakAI, AI Agent Readiness Scanner. https://sitespeak.ai/tools/ai-agent-readiness-scanner, read 2026-11-04.
Last reviewed 2026-11-05