spec-retro
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python helper script to perform text-based churn analysis.
- Evidence:
SKILL.mdincludes instructions for the agent to executepython codex/skills/spec-retro/scripts/spec_churn_detect.pyon user-provided specification and plan files. - Analysis: A thorough review of the script confirms it contains safe, deterministic logic for counting keywords and calculating a 'churn score' using regular expressions. The script does not perform network operations, access sensitive file paths, or use dangerous functions like
eval()orexec(). - [SAFE]: The skill's primary focus is retrospective data synthesis and process improvement, which does not present significant security risks in its current implementation.
Audit Metadata