autoresearch

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the uv package manager installation script from astral.sh, which is a well-known service in the Python development community.
  • [EXTERNAL_DOWNLOADS]: Clones the autoresearch repository from GitHub to establish the training workspace.
  • [REMOTE_CODE_EXECUTION]: Executes the uv installer script using a piped shell command. This is documented as the standard installation procedure for this tool from a recognized service.
  • [COMMAND_EXECUTION]: Utilizes a set of local bash scripts (setup.sh, check-hardware.sh, run-loop.sh) to automate environment preparation and the experiment lifecycle.
  • [COMMAND_EXECUTION]: Invokes uv run train.py to execute model training tasks and extract performance metrics.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 04:58 AM
Security Audit — agent-trust-hub — autoresearch