Use cases

Who uses AgentLitmus, and how

One scanner, four workflows. Whether you ship code, manage client sites, own a brand, or build agents yourself, here's where AgentLitmus fits.

Developers & site owners

Scan before you ship

Catch agent-readiness regressions the same way you'd catch a broken build — before they reach production.

1
Scan your site
Run a scan against staging or production to get a letter grade and a signal-by-signal breakdown.
2
Fix the flagged signals
Each signal links to a concrete fix — missing llms.txt, unrecognized structured data, blocked AI crawlers, and more.
3
Generate llms.txt & robots.txt, add a badge
Generate the files your report is missing, then drop a live grade badge in your README so the score stays visible.

Drop a live grade badge in your README — it always reflects your latest scan.

CI/CD integration via API: coming soon for teams.

Agencies & freelancers

Whether you do web development, SEO, or content strategy, agent readiness is a new axis clients will start asking about.

1
Scan each client site
Get an instant grade and prioritized signal breakdown you can drop straight into an audit deck.
2
Hand clients a grade + fixes
Share the report link directly — no account needed for the client to see exactly what's failing and why.
3
Monitor and get alerted on regressions
Track a site from your dashboard so you hear about a dropped grade before the client does.

Each report is a shareable link with a grade, a full signal breakdown, and concrete fixes — ready to forward to a client as-is.

Content & IP owners / brands

AI agents are already summarizing, citing, and acting on your content. Know what they see — and what you've allowed them to do.

1
See what agents see
A scan shows the same raw content an agent fetches — no JavaScript, no styling, just what's in the response.
2
Control which AI crawlers you allow
Generate a robots.txt that explicitly allows or blocks named AI agents like GPTBot, ClaudeBot, and CCBot, instead of leaving it ambiguous.
3
Check your governance posture
The Adversarial Safety check flags hidden text or injection-style content on your own pages — a governance signal as much as a technical one.

Example: naming AI crawlers explicitly

User-agent: GPTBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: CCBot
Disallow: /

AI & agent builders

AgentLitmus exposes its scanner as an MCP server, so an agent can grade a site, fetch a report, or compare two scans as part of its own workflow.

1
scan_site
Scan a URL and get back a report id — the same checks that power the homepage.
2
get_scan_report / get_domain_history
Fetch a stored report by id, or pull a domain's scan history to see how its grade has trended.
3
diff_scans
Compare two scans of a domain and get a human-readable summary of what changed.

Example: calling scan_site over MCP

{
  "method": "tools/call",
  "params": {
    "name": "scan_site",
    "arguments": { "url": "https://example.com" }
  }
}

The MCP server is available at /api/mcp over Streamable HTTP, with a manifest at /.well-known/mcp.json.