code-smell-validator
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes arbitrary source code files from the user's project to identify code smells. This creates a surface where malicious instructions embedded within the code (e.g., in comments or strings) could potentially influence the agent's logic or the output of the refactoring plan.
- Ingestion points: File contents processed during Phase 2 ('Data Gathering & Analysis') across all scanning scripts.
- Boundary markers: None explicitly mentioned to distinguish between code and embedded instructions.
- Capability inventory: Ability to write refactoring plan files to the local file system and marshal sub-agents for further tasks.
- Sanitization: No explicit sanitization or escaping of the ingested code content is performed before analysis.
- [COMMAND_EXECUTION]: The skill executes several local shell scripts to perform environment detection, git churn analysis, and code hygiene audits. These scripts invoke standard utilities such as 'git', 'grep', and 'npx'.
- Evidence: 'scripts/detect_stack.sh', 'scripts/analyze_churn.sh', 'scripts/check_hygiene.sh', and 'scripts/check_ts_lint.sh' are called during the analysis phases.
Audit Metadata