continuous-learning
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires registering a bash script (
evaluate-session.sh) as a 'Stop' hook in the host application's settings. This script is executed automatically at the end of each session to calculate session length and suggest pattern extraction. - [SAFE]: The provided bash script performs read-only operations on local session transcripts and creates a local directory for learned skills (
~/.claude/skills/learned/). No network requests or data exfiltration attempts were detected. - [SAFE]: The skill does not include any external dependencies or remote code downloads. It relies on standard system utilities like
bash,grep, andjqfor its logic.
Audit Metadata