maintain-verification-skill
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary function is to audit and maintain verification documentation by reading source code and executing local test harnesses. All activities are confined to the local project environment and the skill explicitly limits its edit scope to its own directory, preventing unintended modifications to product code.
- [COMMAND_EXECUTION]: The skill instructions include executing project-local harness scripts and driving application instances for live verification. This is a core part of its functionality and is performed within the project context.
- [PROMPT_INJECTION]: The skill ingests data from project source code and feature files, creating a potential surface for indirect prompt injection. (1) Ingestion points: Project source code files and files within the verification skill's directory (e.g., features/). (2) Boundary markers: The instructions do not specify explicit delimiters or warnings for the agent to ignore instructions embedded in the source data. (3) Capability inventory: The skill has capabilities to read files, create pull requests, and execute local harness scripts via the agent's shell tools. (4) Sanitization: There is no mention of sanitization, validation, or filtering of the content read from external source files.
Audit Metadata