Agentjail
About
What does Agentjail do?
Agentjail offers open-source policy guardrails for coding agents, preventing dangerous tool calls like `rm -rf ~/.ssh` before they execute. It provides granular, deterministic policy enforcement through Rego, serving developers who need to secure their AI agents against unintended or malicious actions. Agentjail ensures that only safe tool calls pass through, with a default-deny stance and auditable decision logs.
What industry does Agentjail operate in?
Agentjail operates in Agentic AI, AI Agent, AI Safety, AI Governance, AI Testing, Developer Tools.
Applied research notes on model internals, representation geometry, attention, and numerical precision.
White Sky develops institutional-grade embodied intelligence, autonomous mission systems, and deployment infrastructure for government, critical infrastructure, and enterprise environments.
Auspere is the security and governance control layer for agentic AI.
Seeing AI from the right angle. Free, open tools, resources, reports, and data that show what AI systems actually do.
Lihuo is developing a runtime constitutional system for generative AI, focusing on governance and responsible AI output.
Something to bend your brain is loading. mindboggling.ai — coming soon.
ShieldTrust is building the security and trust layer for AI that acts on critical systems, enabling control, enforcement, and provable trust for autonomous AI.
AI compliance solutions for EU SMEs, focusing on Article 50 transparency.
An entrant is a company tagged AI Safety whose domain was first registered in the window, counted from registry records in the StartupHub directory. 5 of them registered in the last 30 days. Registry detection runs two to three weeks behind registration, so recent weeks are a floor.
No comments yet. Be the first to share your take.