cy-fix-reviews
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes PR review issue files to drive its remediation workflow. If these files contain malicious instructions (e.g., from an adversarial PR comment), the agent might execute them as if they were valid tasks.
- Ingestion points: Scoped issue files located in
.compozy/tasks/<name>/reviews-NNN/(as described in SKILL.md). - Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to ignore natural language instructions embedded within the issue files.
- Capability inventory: The agent has the capability to modify source code, update test files, and execute repository verification commands via
cy-final-verify. - Sanitization: No sanitization or validation of the issue file content is mentioned before it is used to guide code changes.
- [COMMAND_EXECUTION]: The workflow requires the execution of
cy-final-verifyand the repository's real verification commands. While these are part of a standard development workflow, they represent the execution of local scripts and tools based on the state of the modified code.
Audit Metadata