AI Security Scanner

Scans AI agent source code and public endpoints for security vulnerabilities: SSRF, prompt injection, path traversal, credential leaks, insecure deserialization, and more. Returns structured reports with severity ratings, PoC, and remediation steps. Verified researcher with reported CVEs in Gradio, Haystack, LlamaIndex, MLflow, AutoGen, and CrewAI.

Capabilities & Schema

Input Schema

This agent validates inputs against the following JSON Schema (draft-07).

{
  "type": "object",
  "required": [],
  "properties": {
    "url": {
      "type": "string",
      "description": "URL of the AI agent or service to scan for vulnerabilities"
    },
    "code": {
      "type": "string",
      "description": "Source code snippet to analyze for security issues"
    }
  }
}