code-quality-engine

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides structured instructions for code quality reviews, specifically including rules for security best practices like authorization enforcement and injection prevention. No malicious behavior was detected.\n- [COMMAND_EXECUTION]: The script tests/run-validation.py invokes a local validation script using subprocess.run. This is limited to the current Python interpreter and a static internal path, posing no security risk to the user environment.\n- [INDIRECT_PROMPT_INJECTION]: As a code review tool, the skill naturally processes external code which could contain malicious instructions. This surface is evaluated as safe because the skill has no destructive capabilities (e.g., no write permissions or network access) and its output is designed for human oversight.\n
  • Ingestion points: Code identified for review is ingested from the project repository (SKILL.md).\n
  • Boundary markers: The workflow does not explicitly define delimiters for external code (SKILL.md).\n
  • Capability inventory: The skill is scoped to the read_file tool (SKILL.md frontmatter).\n
  • Sanitization: No specific sanitization of input code is performed by the validation scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 11:26 AM