Resources

Every integration built on OGR — agent hooks, gateway hooks, sandbox hooks, and the eBPF sensor — with install hints and source links.

Operational guides and notes on guarding real agents: Claude Code, opencode, OpenClaw, Hermes, and a real malware incident.

GitHub, Discord, contributing, governance, and how to add your own integration or detector.

The neutral detector benchmark: shared corpora, an open harness, and a leaderboard. Detectors compete, we referee.

GuardEvent, Verdict, provenance, composition, attestation, degraded mode, and the Runtime API binding — the normative sources.

OGR is designed to be installed and operated by an agent: draft a policy, get human approval, enforce. The skill packages the whole procedure.