diffusion-release-tracking

Pass

Audited by Gen Agent Trust Hub on May 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, data exfiltration, or unauthorized command execution detected. The skill is purely instructional, guiding the agent to collect and evaluate product metrics.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests untrusted user signals (e.g., feedback, metrics). However, the risk is negligible as the skill's capabilities are limited to generating text-based reports and recommendations without performing sensitive system operations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 21, 2026, 08:11 AM
Security Audit — agent-trust-hub — diffusion-release-tracking