unsafe-checker

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill is designed to ingest and analyze untrusted external Rust source files via 'Read', 'Grep', and 'Glob' tools. Evidence Chain: 1. Ingestion point: Skill targets all '.rs' files via the specified glob. 2. Boundary markers: No markers or 'ignore' instructions are provided to delimit untrusted code from agent instructions. 3. Capability inventory: Access is restricted to read-only file operations (no execution or network capabilities). 4. Sanitization: None. Although the skill lacks dangerous side effects, malicious comments in processed code could theoretically influence the agent's reasoning or summary of that code.
  • [General Security] (SAFE): No malicious instructions, obfuscation, persistence mechanisms, or unauthorized data access patterns were identified in the skill content or metadata.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 05:07 AM