qa-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from various project files including source code and configuration documents (e.g., AGENTS.md, CLAUDE.md). This creates a surface where instructions embedded in project data could influence agent behavior during a review pass. \n
- Ingestion points: Reads local project files, requirement checklists, and deliverable summaries (SKILL.md, full-guide.md). \n
- Boundary markers: None explicitly defined to isolate external data from agent instructions. \n
- Capability inventory: Shell commands (ls, grep, find, git) and Python execution (python3 -c). \n
- Sanitization: No explicit content sanitization or instruction-bypass protection is mentioned. \n- [DYNAMIC_EXECUTION]: The skill utilizes dynamic execution by suggesting the generation of small Python scripts to perform specific validation tasks, such as verifying that markdown code blocks are properly balanced. \n
- Execution method: Invokes python3 -c to run logic directly on file contents (full-guide.md). \n- [COMMAND_EXECUTION]: The workflow relies on executing standard operating system utilities to perform its verification duties. \n
- Evidence: Uses ls, grep, find, and git commands to audit the state of the codebase and verify the presence of required files and patterns (SKILL.md, full-guide.md).
Audit Metadata