refactor-research-code

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a local Python script scripts/release_audit.py to analyze the target codebase. Technical review of the script confirms it is read-only, uses only Python standard library modules (re, pathlib, json), and performs no network operations or file system modifications.
  • [COMMAND_EXECUTION]: The skill uses ../reflect-and-improve/scripts/reflect_log.py to update a cross-skill memory log. This interaction is restricted to logging recurring issues and recommendations for future sessions.
  • [SAFE]: The skill implements a robust 'prime directive' for result preservation, strictly separating 'SAFE' mechanical cleanups from 'ASK-FIRST' changes that could affect numerical output. This methodology prioritizes research integrity and enforces human-in-the-loop confirmation.
  • [SAFE]: The skill mitigates risks associated with processing untrusted research code by using a deterministic static audit script and requiring external verification (re-running the pipeline) rather than relying solely on the LLM's interpretation of changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 01:03 PM
Security Audit — agent-trust-hub — refactor-research-code