api-schema-validation
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data sources including OpenAPI specifications, JSON schemas, GraphQL schemas, and request-response samples which could contain embedded malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill reads and analyzes external API schemas, endpoint inventories, and request-response samples as specified in
SKILL.mdandprompts/api-schema-validation.md. - Boundary markers: The instructions lack specific delimiters or isolation techniques (such as XML tags or explicit 'ignore instructions' wrappers) for the ingested API material to prevent the agent from following instructions hidden within the data.
- Capability inventory: The skill generates text-based findings (Markdown, JSON, CSV) and does not explicitly call dangerous tools; however, it is designed to operate within an agent environment that typically provides shell and file system access.
- Sanitization: There is no evidence of sanitization, filtering, or validation of the input schema or sample content to prevent malicious payload interpolation.
Audit Metadata