retro
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Python script (
extract_conversation.py) to parse local Claude Code JSONL logs. The commands executed are localized to the user's home directory (~/.dotfiles/claude.symlink/skills/retro/) and operate on local data only. The script uses Python standard libraries exclusively. - [DATA_EXPOSURE]: The skill accesses Claude Code session logs located in
~/.claude/projects/. While these logs contain conversation history, the script is designed to filter out sensitive tool outputs (like file contents or command results) and only extract human-readable prose for the purpose of a retrospective. No data is sent to external servers.
Audit Metadata