pm-spec-quality-audit

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill consists of instructional text designed to guide the AI in performing a specialized document review. No evidence of prompt injection, obfuscation, or privilege escalation was found.
  • [EXTERNAL_DOWNLOADS]: The skill includes several links to educational lessons on uxcel.com. These are legitimate references to the author's own domain (Uxcel-Lab) used for documentation and context, presenting no security risk.
  • [DATA_EXFILTRATION]: No network operations, credential access, or sensitive file paths were detected within the instructions.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is intended to process external documents (product specs).
  • Ingestion points: Product specifications, PRDs, or requirements documents provided by users (SKILL.md).
  • Boundary markers: None present; the instructions do not define delimiters to isolate untrusted data.
  • Capability inventory: Analysis and text generation only. No file system access, network calls, or code execution capabilities are present.
  • Sanitization: None present.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 01:50 PM
Security Audit — agent-trust-hub — pm-spec-quality-audit