autoresearch
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes agent-generated bash scripts and various git commands to automate the process of modifying code, running benchmarks, and reverting unsuccessful changes.
- [SAFE]: No external network communication, remote script downloads, or credential exposure patterns were found in the skill logic.
- [SAFE]: The 'autoresearch_helper.py' script relies on Python standard libraries for metric calculation and local JSONL file management.
Audit Metadata