cleanse
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute project-specific lint or build commands to verify code integrity after simplification. This capability allows the agent to run arbitrary shell commands based on its inference of the project's environment.
- [PROMPT_INJECTION]: The skill's functionality involves reading and processing untrusted source code, which serves as an attack surface for indirect prompt injection. Instructions embedded in comments or metadata within the processed files could potentially be interpreted as commands by the agent.
- Ingestion points: Target files or directories provided by the user (SKILL.md).
- Boundary markers: Not explicitly used to isolate instructions from the source code data.
- Capability inventory: Includes the ability to modify local files and execute shell commands for build/lint verification.
- Sanitization: No sanitization or validation of the input source code is performed prior to processing.
Audit Metadata