ExtractWisdom

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses fabric for YouTube transcripts and WebFetch for articles, which are standard tools for its stated purpose of wisdom extraction.
  • [SAFE]: The skill includes instructions to check for local user customizations in ~/.claude/LIFEOS/USER/CUSTOMIZATIONS/SKILLS/ExtractWisdom/ and logs execution metadata to ~/.claude/LIFEOS/MEMORY/SKILLS/execution.jsonl. These are standard patterns for personalized AI agent skills and do not involve sensitive credential access or exfiltration.
  • [SAFE]: No prompt injection, obfuscation, or unauthorized privilege escalation patterns were detected. The instructions focus on tone, voice, and quality of the generated summary.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 08:46 AM
Security Audit — agent-trust-hub — ExtractWisdom