Cloudflare Security Audit Skill
Multi-phase security audits with verified, machine-readable findings
Policy checkpoint that audits and blocks risky agent actions
Costa Score is an open-source governance layer that checks every AI agent action against policy before it executes. It is for teams handling sensitive data that need to block PII leaks, prompt injection and risky network calls with an audit trail.
Before an agent acts, the requested action is evaluated against configured policies and either allowed or blocked. Every decision is written to a tamper-evident audit trail so teams can review what happened and why.
This week’s repos: PS5 game on a PC! Turn agents into employees. AI video maker.
Multi-phase security audits with verified, machine-readable findings
Routes coding agents to the right reverse-engineering tools
Self-hosted Bitwarden-compatible password manager server
Independent audit for AI agents in under two minutes
Kernel-level sandbox that decides what agents may touch
Kubernetes security scanner that finds and fixes misconfigurations