domains / families / agent-tooling / harness-audit

harness-audit

separate

Audit an agent's harness against the four rings: containment, guides, sensors, permissions. Use when the user asks whether their agent setup is safe, wants a review of sandboxing / permissions / hooks / AGENTS.md guardrails, or asks what a bad session could break. Read-only: reports the blast radius and a hardening list, changes nothing. Do NOT use for context layout (context-audit) or cost routing (gate-check).

When to use

• reviewing agent setup safety
• auditing sandboxing and permissions
• checking hooks and AGENTS.md guardrails
• assessing what a bad session could break
• reporting blast radius
• generating a hardening list

When not to use

• context layout
• cost routing
• applying fixes
• hardening the harness

Skill page

Source