harness-bootstrap-brownfield

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions explicitly mandate read-only access to the source code, preventing any unauthorized modifications during the analysis process.
  • [SAFE]: Output operations are strictly limited to the {product_root}/docs/context/ directory, which protects the rest of the filesystem from arbitrary write operations.
  • [SAFE]: The identification of the project stack relies on analyzing standard manifest files such as package.json and composer.json. It looks for .env.example as a stack signal but does not attempt to read sensitive .env files containing production secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 08:16 PM
Security Audit — agent-trust-hub — harness-bootstrap-brownfield