harness-engineering-playbook

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The harness_wizard.py script executes internal shell scripts (bootstrap_harness.sh, audit_harness.sh) using subprocess.run. This is used to perform repository initialization and auditing tasks as part of the skill's core functionality.
  • [COMMAND_EXECUTION]: Several provided shell templates (lint.sh, smoke.sh, test.sh, typecheck.sh) use eval to execute shell commands specified in environment variables (e.g., HARNESS_LINT_CMD). This enables flexible command customization by the user but involves dynamic execution of shell strings.
  • [EXTERNAL_DOWNLOADS]: The documentation references the author's repository and suggests using npx to install the skill from the broomva/skills monorepo, which is a standard installation pattern.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 09:56 AM
Security Audit — agent-trust-hub — harness-engineering-playbook