swift-clean-types

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill uses a direct interpolation pattern for user-provided input without boundary markers or instructions to disregard embedded commands.
  • Ingestion points: The $ARGUMENTS variable is directly placed into the instruction block in SKILL.md.
  • Boundary markers: None. The input is not wrapped in delimiters (like XML tags or markdown blocks) that would help the agent distinguish between user data and system instructions.
  • Capability inventory: The skill is configured with Read, Edit, Glob, Grep, and Bash tools in its frontmatter, providing significant file system and shell access.
  • Sanitization: No validation or sanitization is performed on the input to ensure it is a valid file path or code snippet before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 01:40 PM
Security Audit — agent-trust-hub — swift-clean-types