ai-slop-cleaner

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues were identified. The skill defines a disciplined refactoring methodology that prioritizes behavior preservation and systematic verification.- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to process untrusted code files. Ingestion points: User-provided code files and changed-files lists (SKILL.md). Boundary markers: Absent; there are no instructions to delimit or ignore instructions within the processed data. Capability inventory: Execution of regression tests, linting, typechecking, and security scans (SKILL.md). Sanitization: Absent; the skill does not include steps to sanitize or escape file content before processing. This constitutes a standard vulnerability surface for developer tools, which is addressed here by the skill's reliance on verification gates.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 05:06 PM