Open-source security firewall for AI agents — validates tool calls, strips ghost arguments, enforces type safety, PII masking, RBAC, cost tracking & sandbox isolation. Works with LangChain, OpenAI Agents SDK, PydanticAI & CrewAI.

A secure, rootless Podman sandbox for running AI coding agents (Antigravity, Cursor) with SSH forwarding and full host isolation.

A secure “airlock” for AI agents to interact with APIs using risk-based decisions, Auth0 Token Vault, and human-in-the-loop approvals.

Credential-isolating reverse proxy for AI agents. Lets agents call APIs without seeing the keys.

airlock is a cryptographic handshake protocol for verifying AI model identity at runtime. It enables real-time attestation of model provenance, environment integrity, and agent authenticity - without relying on vendor trust or static manifests.

OpenClaw Airlock: A secure authorization gateway for AI agents, enforcing risk-tiered policies, human-in-the-loop approvals, delegated token exchange, and full audit logging. Enables safe automation across GitHub, Gmail, Slack, and other APIs while maintaining governance and trust.

a good fence between you and your coding agents

A trust boundary between AI agents and infrastructure. Agents write code, Airlock runs it with injected secrets.

Agent skills, component library, UI templates, moodboard, and template intake pipeline for Airlock

Filesystem tools for agents. Read, write, replace, grep, list, create, delete, move via VFS airlock.

A robust compliance layer that acts as a secure airlock for AI Agents, preventing sensitive data leaks via reversible tokenization. Built with strict typing and a custom Python backend to bridge the gap between innovation and data privacy.