# AgentCapable — can AI agents actually reach your site?

ChatGPT, Claude and Perplexity send agents to read websites millions of
times a day. If your firewall quietly turns them away, you disappear from AI
answers — and you can't see it from the inside. One scan shows you exactly
what happens when 14 different AI agents knock on your door.

Free full report. A fresh measurement every time. Takes about a minute.

## What you'll get

- **A straight answer** — one verdict (Open / Partially blocked /
  Agent-hostile / Contradictory / Unknown) with the evidence behind it. When
  something can't be measured, the report says so instead of guessing.
- **Agent-by-agent detail** — which of the 14 agents got your page, which
  were turned away, and which ones your robots.txt promised to let in.
- **A score with exact fixes** — a 0-100 readability score built from 23
  named checks, each with what we saw, why it matters, and a paste-ready fix.

## How it works

1. **We knock** — your site requested the way 14 real AI agents request it,
   from two independent datacenter networks, with a normal-browser baseline.
2. **We compare** — what robots.txt declares per agent vs what your edge
   actually did; then 23 checks on how readable the site is once an agent is in.
3. **We judge honestly** — a finding counts only when both networks agree;
   anything unmeasurable is excluded from the score, never counted against you.

## What we will never claim

Our requests come from datacenter networks — where AI agents actually run —
but we cannot send traffic from the AI vendors' own verified addresses. The
strongest claim a report makes is that your edge treats agent requests from
cloud servers as hostile, never that a specific vendor's bot is blocked.
Reports are never stored; every scan measures the site as it is right now.

## API

POST https://agentcapable.dev/scan with
{"domain": "yourdomain.com", "lead": {"name": "...", "email": "...", "phone": "..."}}

More: [llms.txt](https://agentcapable.dev/llms.txt) ·
[Scanner policy](https://agentcapable.dev/bot)