idor-testing
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external web applications, such as API responses and object identifiers, to detect vulnerabilities. This creates an attack surface where malicious data could influence agent behavior.
- Ingestion points: Target web application URLs and responses from API endpoints (SKILL.md).
- Boundary markers: The skill specifies a mandatory confirmation gate requiring user approval for targets and commands.
- Capability inventory: Includes capabilities for making HTTP requests and performing automated parameter enumeration.
- Sanitization: Lacks specific instructions for sanitizing or filtering data received from external sources.
Audit Metadata