spec-probes

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is a purely instructional framework for software specification review. It contains no executable scripts, binaries, or system commands.- [SAFE]: No external network requests, package dependencies, or remote code patterns were detected.- [SAFE]: No obfuscation, prompt injection, or data exfiltration vectors are present in the documentation or metadata.- [SAFE]: While the skill processes external input (user requirements), it lacks the capabilities (such as file writing or network access) required to manifest an indirect prompt injection attack.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:55 PM
Security Audit — agent-trust-hub — spec-probes