pegasi-ai/reins
A security control layer for AI agents that enforces deterministic security policies on agent actions.
Not currently ranked — collecting fresh signals.
star history
Reins provides security controls for AI agents by enforcing deterministic security policies on every agent action. It scans configurations for OWASP ASI10 vulnerabilities, tracks drift over time, and evaluates policies in under 50ms. The tool integrates with Claude Code PreToolUse and PostToolUse hooks, OpenClaw, and any MCP-compatible agent, serving as a developer toolkit for agent security and observability.
Frequently asked
- What is pegasi-ai/reins?
- A security control layer for AI agents that enforces deterministic security policies on agent actions.
- Is reins open source?
- Yes — pegasi-ai/reins is open source, released under the Apache-2.0 license.
- What language is reins written in?
- pegasi-ai/reins is primarily written in Python.
- How popular is reins?
- pegasi-ai/reins has 408 stars on GitHub.
- Where can I find reins?
- pegasi-ai/reins is on GitHub at https://github.com/pegasi-ai/reins.