unsafe-checker

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is comprised of markdown instructions and Rust code examples designed for educational and auditing purposes. No functional scripts or executables are present within the skill folder.
  • [SAFE]: Tool access is strictly limited to 'Read', 'Grep', and 'Glob', which ensures the agent can only analyze project files without the ability to modify them or send data over a network.
  • [SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was detected. All code snippets provided as 'BAD' patterns are clearly documented as anti-patterns for review reference.
  • [SAFE]: The skill possesses an indirect prompt injection surface as it analyzes user-provided Rust files (**/*.rs). Ingestion occurs via 'Read' and 'Grep' tools. While boundary markers and sanitization are absent, the security risk is limited by the restricted capability set (read-only, no network access).
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 05:49 PM
Security Audit — agent-trust-hub — unsafe-checker