auto-approve-security

Securing autonomous workflows with auto-approve enabled requires documenting the threat model and applying explicit deny rules. This Skill flags security gaps and provides PreToolUse hook constraints to prevent arbitrary code execution.

Common exploit chains in auto-approve permission systems include edit-and-execute attacks, script injection, git hooks, and configuration poisoning. This Skill provides step-by-step analysis of these attack paths and their mitigations.

Configuring deny and allow rules involves setting explicit PreToolUse hook constraints and filesystem protections for settings and hooks. This Skill provides concrete rule recommendations to block exploit chains and configuration poisoning.

Yes, this threat model analysis covers build and test command execution, script execution, git hooks, and shell interactions. It is designed specifically for autonomous workflows where Edit and Write are auto-approved.

Auditing auto-approve security rules involves running practical tests against documented exploit chains to validate deny rules and hooks. This Skill provides an audit checklist with testing notes and verified outcomes for real sessions.

Preventing configuration poisoning in auto-approved environments requires applying filesystem protections for settings, hooks, and skills. This Skill documents the threat model and provides concrete mitigations against configuration poisoning attacks.