api-fuzzing
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and parse untrusted external data, specifically OpenAPI/Swagger specifications, Postman collections, and JavaScript bundles.
- Ingestion points: The 'Method' section details harvesting endpoints from
swagger.json,/api-docs, and code bundles. - Boundary markers: There are no instructions provided to the agent to treat external specification content as non-instructional or to use specific delimiters.
- Capability inventory: The skill leverages network-interactive tools (
ffuf,kiterunner,arjun) to perform operations based on the harvested data. - Sanitization: No explicit sanitization, validation, or schema enforcement is mentioned for the external files prior to processing.
Audit Metadata