Show HN: Paveo – Check Claude Code, Codex and Cursor commands before they run

Category: security

Tags: ai-agents, security, policy-engine, audit-logging, coding-agents

Score: 7.0/10 (Innovation: 6, Technical: 7, Documentation: 8, Utility: 7)

Paveo is a policy enforcement layer that sits between AI coding agents (Claude Code, Codex, Cursor) and the actions they take, blocking dangerous commands and file writes before they execute. It also guards custom agents and MCP servers, logging every decision in a tamper-evident hash chain. The project addresses a real gap in AI agent safety and is notably strict about local-only operation and supply-chain integrity.

Target audience: platform engineers, security teams, AI agent developers, devops

Repository: https://github.com/paveo-dev/paveo · Python · Apache-2.0 · 3 stars

View on Hacker News