Fix the score for crisp.chat

Currently: 20 · target: 95+

Here's exactly what we'd ship for crisp.chat:

+25llms.txt — agent-facing site summary

# /llms.txt for crisp.chat
> One-line description of crisp.chat for AI agents.

## Key pages
- /: homepage
- /api: API docs (if any)

+20.well-known/ai-plugin.json — ChatGPT/Claude manifest

{
  "schema_version": "v1",
  "name_for_model": "crisp.chat",
  "description_for_model": "What crisp.chat does, for an AI agent.",
  "api": { "type": "openapi", "url": "https://crisp.chat/openapi.yaml" }
}

+20openapi.yaml — public endpoints in OpenAPI 3

✓/api/v1 — literal JSON index for agents — already present

+10MCP server — first-class agent tool access

+5robots.txt — explicit AI-crawler allow rules

# robots.txt for crisp.chat — allow AI agents
User-agent: GPTBot
Allow: /
User-agent: ClaudeBot
Allow: /
User-agent: PerplexityBot
Allow: /
Sitemap: https://crisp.chat/sitemap.xml

✓Schema.org JSON-LD — Organization/FAQ markup — already present

Projected: 20 → 100 after the 7-signal uplift

Two ways to fix it: get the $29 report instantly on-screen — the exact files & fixes to apply yourself, no email or repo needed — or have us do it for $199, delivered as a pull request within 72 hours. $199 tier: full refund if your score doesn't hit 90+.

← Back to score report