qa-full
Warn
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to automatically infer and execute test and build commands from the project's configuration files (e.g., package.json, Makefile, pyproject.toml). This results in the execution of arbitrary shell commands defined by the repository being analyzed.
- [DYNAMIC_EXECUTION]: The skill dynamically identifies and executes commands based on the structure and content of the target project, specifically within Step 2 (Tests & Build) and when triggering sub-skills like /db-optimize or /web-perf.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data that can influence its operational behavior.
- Ingestion points: The skill reads and obeys configuration instructions provided in the project's root CLAUDE.md file (Step: Project configuration). It also processes branch diffs and test logs.
- Boundary markers: The instructions do not specify any boundary markers or ignore-directives for embedded instructions in the ingested files.
- Capability inventory: The skill utilizes powerful tools including Bash, Task, and WebFetch, and performs file-writing operations to record reports.
- Sanitization: No sanitization or validation logic is defined for the instructions parsed from CLAUDE.md; the agent is explicitly instructed to honor the file's content.
Audit Metadata