rule
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions focus on standard development workflow automation, such as reading configuration files and updating source code for consistency. No evidence of credential harvesting, unauthorized network activity, or malicious code execution was found.
- [INDIRECT_PROMPT_INJECTION]: The skill reads from potentially untrusted project files and source code to identify rule violations, which constitutes a vulnerability surface. However, the risk is inherent to the skill's primary function of linting and refactoring, and the instructions are limited to structured pattern matching.
- Ingestion points: Reads instructions and patterns from
AGENTS.md,CLAUDE.md, and local code files. - Boundary markers: None specified.
- Capability inventory: Reads and writes to the local filesystem to modify project documentation and source code.
- Sanitization: No sanitization logic for input data is defined.
Audit Metadata