Skip to content
MoorAI
// technology

How it works

One check where the AI acts, before it acts. It runs on the device or next to the workload, and the console gets signals, never your prompts, code or files.

// how it works

One local check, on the device, before a prompt ever reaches the AI — then only signals, never the prompt, flow to the console.

The check is a hook the agent has to pass through before every tool call runs — not an MCP tool the model may choose not to call. Enforcement is validated on Claude Code.

One check, on the device — before the AI sees anything. // moorai · on-device AI guardrails Employee types a prompt MoorAI 79-threat matrix · content rules · local coach alert block justify prompt content never reaches the console AI agent Claude · Codex Copilot CLI Model API cloud LLM Central console posture · inventory · compliance — content-free signals only // four pillars On-device DLP Prompts and outputs checkedlocally against policy — nothinghas to leave the endpoint. AI-SPM for endpoints Shadow-AI discovery plusagent, tool & MCP governancewhere the work happens. Human-in-the-loop Coach, alert, block, or requirea signed justification — setper policy, tenant, device. Continuous compliance OWASP LLM · MITRE ATLASNIST AI RMF · ISO 42001EU AI Act-aware.
// go deeper

The layers, and where they run.

// workflow
Five layers, three moves
Decode, detect, escalate, enforce and record, all on the device.
 // endpoints
MoorAI for Endpoints
Coding agents and everyday AI use on the devices people work on.
 // runtime
MoorAI for Runtime
The AI apps, agents, RAG pipelines and APIs you ship.

Check it
where the AI acts.

Free and open source (MIT). No account to start.

Download MoorAI Quick start Community agent on GitHub