skill-auditor

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill operates as a read-only auditor, using platform tools to examine repository metadata. No malicious behaviors, obfuscation, or unauthorized data access were identified. The hardcoded local path in the documentation (/Users/ravishankar/...) is a metadata reference and does not represent a sensitive file access attempt.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as its primary function involves processing untrusted content from other skill files to generate a summary report.
  • Ingestion points: Reads file content from the skills/ directory via Glob and Read tools.
  • Boundary markers: None; the skill lacks specific markers to distinguish between audited content and its own instructions in the output.
  • Capability inventory: Read, Glob, Grep, Bash (used for line counting and directory listing only).
  • Sanitization: Extracted text is reported without escaping or filtering for potential injection patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 01:19 PM
Security Audit — agent-trust-hub — skill-auditor