api-security-testing
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for running the
strixcommand-line tool. This includes passing user-supplied target URLs, local file paths for schemas (OpenAPI, Postman collections), and configuration flags to the tool. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, potentially untrusted data sources to define its scanning scope.
- Ingestion points: External data enters the agent context through API schemas (OpenAPI/Swagger, GraphQL introspection, Postman collections) and gRPC proto files passed via the
-tand--workspace-fileflags inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or "ignore instructions" headers for the content of the imported schemas.
- Capability inventory: The
strixtool has the capability to perform arbitrary network requests to enumerated API endpoints and write scan results/vulnerability reports to the local file system (e.g.,strix_runs/,vulnerabilities/). - Sanitization: The skill relies on the
strixtool's internal parsing logic; no additional sanitization or validation of the input schemas is specified in the prompt instructions.
Audit Metadata