A policy-based security gate that blocks unsafe crypto transactions and auto-fires remediation playbooks with tamper-evident audit
WCP — Worker Class Protocol specification
NORNR Sentry local airbag for dangerous agent actions
Authorized CI/CD auto-pentest policy gate for GitHub Actions
Paid remote MCP for AI agent safety replay checks, policy gates, eval receipts, control-fix suggestions, and release evidence exports.
Report-only pre-merge policy gate as a Claude Code skill — checks a diff against versioned rulesets and returns a PASS/WARN/FAIL verdict. Never edits code.
TypeScript coding-agent runtime showcase with policy-gated tools, audit logging, secret redaction, fake/OpenAI-compatible providers, and plugin loading.
Local deterministic intake gate for instruction-bearing repo surfaces with explicit human acknowledgment before approval.
Local deterministic CLI that checks proposed agent actions against approved scope and flags when human reapproval is required.
SLSA/SBOM Evidence Bridge Layer connecting SBOM, provenance, artifact attestation, policy gate, and Ed25519 review evidence to the REMEDA/QSP verification rail.
PyHall Python SDK — Python reference implementation of WCP (Worker Class Protocol)
Safeloop is a lightweight governance SDK for local AI agent loops: policy gate, circuit breaker, and action ledger.
Add a description, image, and links to the policy-gate topic page so that developers can more easily learn about it.
To associate your repository with the policy-gate topic, visit your repo's landing page and select "manage topics."