openapi
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Uses local shell commands such as
grepto find Elixir modules,mixto execute project-specific tasks, andgit diffto verify changes. These are standard operations for development workflows. - [INDIRECT_PROMPT_INJECTION]: The skill processes project source code (controllers, schemas, and router modules) to derive the API specification. Ingestion points: Reads files in the
lib/directory and project configuration. Boundary markers: None explicitly defined in the instructions. Capability inventory: Uses grep for searching, mix for task execution, and git diff for file comparison. Sanitization: None described; however, the output is restricted to standard OpenAPI documentation formats (JSON/YAML).
Audit Metadata