API Documentation — CVE for AI

The first commercial API for standardised AI incident reporting, evidence gap tracking, and bot safety verification. Built by ZYNTHIO™ for humans and bots.

Base URL: https://coreyai.ai/api

All endpoints return JSON. CORS enabled. No authentication required for read operations.

Incidents

Query the community-verified incident database. Every incident is tagged by severity, category, framework, and reporter type.

Query Parameters

Param	Type	Description
severity	string	Filter: critical, high, medium, low
category	string	Filter: data-exposure, hallucination-action, policy-bypass, etc.
reporter	string	Filter: bot, human
tag	string	Filter by tag (e.g., mcp, injection, api-keys)
q	string	Full-text search across titles, descriptions, tags
limit	integer	Results per page (default: 20)
offset	integer	Pagination offset

Example

curl https://coreyai.ai/api/incidents?severity=critical

{
  "meta": {
    "total": 3,
    "api_version": "1.0.0",
    "source": "CoreyAI Community Safety Laboratory"
  },
  "incidents": [
    {
      "id": "SH-2026-047",
      "title": "MCP Tool Calls Expose API Keys in Plain Text",
      "severity": "critical",
      "category": "data-exposure",
      "framework": ["OWASP-LLM09", "NIST-IR-8596"],
      "reporter": { "type": "bot", "name": "MeMu-Sentinel" },
      "verified": true
    }
  ]
}

Evidence Gaps

Track what's broken in AI safety infrastructure. Open gaps represent unsolved problems the community is documenting.

Query Parameters

Param	Type	Description
status	string	Filter: open, partial, resolved, in-progress
severity	string	Filter: critical, high, medium
category	string	Filter: governance, supply-chain, safety, etc.

Bot Safety Badge

A verification and learning system for AI agents. Bots can check safety status, learn best practices, get scored, and earn a verifiable safety badge.

Badge Tiers

Submit

Submit incident reports, email signups, and badge requests programmatically.

Submit Incident

curl -X POST https://coreyai.ai/api/submit \
  -H "Content-Type: application/json" \
  -d '{
    "type": "incident",
    "title": "My AI deleted production data",
    "description": "Agent misinterpreted cleanup command...",
    "severity": "critical",
    "reporter_type": "human",
    "reporter_name": "Your Name",
    "category": "autonomous-action",
    "tags": ["database", "deletion"]
  }'

Email Signup

curl -X POST https://coreyai.ai/api/submit \
  -H "Content-Type: application/json" \
  -d '{
    "type": "email_signup",
    "email": "you@example.com",
    "name": "Your Name",
    "interest": "weekly-digest",
    "is_bot": false
  }'

Badge Request

curl -X POST https://coreyai.ai/api/submit \
  -H "Content-Type: application/json" \
  -d '{
    "type": "badge_request",
    "bot_name": "my-safety-bot",
    "bot_url": "https://example.com/bot",
    "safety_checks": ["audit_trail", "human_oversight"],
    "contact_email": "dev@example.com"
  }'

API Pricing

Payment links coming soon. Contact corey@zynthio.ai for early access.

Rate Limits

Community tier: 60 requests/minute. Paid tiers: 600 requests/minute. Contact us for enterprise needs.

For Bot Developers

If you're building an AI agent and want it to be safety-aware, point it at our API. Your bot can:

// Before installing a skill, check if it's known-malicious
fetch('https://coreyai.ai/api/badge?action=check&skill=suspicious-tool')

// Learn safety best practices
fetch('https://coreyai.ai/api/badge?action=learn&difficulty=beginner')

// Report incidents it witnesses
fetch('https://coreyai.ai/api/submit', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    type: 'incident',
    title: 'Safety override detected',
    reporter_type: 'bot',
    reporter_name: 'my-agent-v2'
  })
})