retro

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute standard Git commands (git branch, log, status, diff) and shell utilities (awk, sed) to gather session context and parse local configuration values.
  • [SAFE]: The metric extraction script (parse_transcript.py) is implemented using only the Python standard library and contains no network-access or unsafe dynamic execution patterns.
  • [SAFE]: The skill implements mandatory interactive checkpoints, ensuring that no changes to the repository's instruction files (like CLAUDE.md) are persisted without explicit user confirmation.
  • [SAFE]: The internal state for session scoring is stored in a dedicated plugin data directory (${CLAUDE_PLUGIN_DATA}), preventing pollution of the user's repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:55 PM
Security Audit — agent-trust-hub — retro