Model-agnostic security runtime for AI coding agents with bounded authority, verified evidence, independent review and human gates.

1 stars 0 forks 1 watchers JavaScript Apache License 2.0
agent-runtime agentic-ai ai-agents ai-safety autonomous-agents claude codex coding-agents human-in-the-loop llm mcp nodejs opencode security windows
3 Open Issues Need Help Last updated: Aug 31, 2026

Open Issues Need Help

View All on GitHub

Model-agnostic security runtime for AI coding agents with bounded authority, verified evidence, independent review and human gates.

JavaScript
#agent-runtime#agentic-ai#ai-agents#ai-safety#autonomous-agents#claude#codex#coding-agents#human-in-the-loop#llm#mcp#nodejs#opencode#security#windows

Model-agnostic security runtime for AI coding agents with bounded authority, verified evidence, independent review and human gates.

JavaScript
#agent-runtime#agentic-ai#ai-agents#ai-safety#autonomous-agents#claude#codex#coding-agents#human-in-the-loop#llm#mcp#nodejs#opencode#security#windows

Model-agnostic security runtime for AI coding agents with bounded authority, verified evidence, independent review and human gates.

JavaScript
#agent-runtime#agentic-ai#ai-agents#ai-safety#autonomous-agents#claude#codex#coding-agents#human-in-the-loop#llm#mcp#nodejs#opencode#security#windows