lint
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external codebase files and executes local linting and testing tools based on the project context. This creates a surface where a malicious repository could include poisoned configuration files (e.g., linter configs, test runners) or source code containing hidden instructions designed to influence the agent during the linting, verification, or commit phases.
- Ingestion points: The agent reads source code and configuration files across the codebase to identify tools and check standards (SKILL.md, Phase 2).
- Boundary markers: No specific delimiters or instructions to ignore embedded content within processed files are provided.
- Capability inventory: The agent is directed to run command-line tools ("run available linting tools", "run tests"), modify files ("Apply auto-formatting"), and create git commits (Phase 4).
- Sanitization: There is no evidence of sanitization or isolation for the output of external tools before the agent processes them for the final report.
Audit Metadata