maintain-verification-skill
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the project's source code and feature files, which could contain instructions designed to influence the agent's behavior.
- Ingestion points: Product source code and project-local feature files.
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the analyzed source code.
- Capability inventory: The agent can execute shell commands, write files within the skill directory, and create Pull Requests.
- Sanitization: No specific filtering or escaping of content read from source files is required before processing.
- [DYNAMIC_EXECUTION]: The skill permits the modification and subsequent execution of local harness scripts, which is a form of dynamic code generation used for verification.
- Evidence: Step 5 instructs the agent to fix harness gaps and document their invocation, which are then run in Step 4.
- [COMMAND_EXECUTION]: The skill relies on executing external shell commands, specifically harness scripts and diagnostic health checks, to verify application behavior.
- Evidence: Step 4 details the invocation of these scripts across different application states and sessions.
Audit Metadata