Govern what your AI does, at runtime.

Coding Agent Hooks

On Claude Code, Gemini CLI and Cursor IDE a deny rule stops the call before it runs, across every tool family, and the same holds on Kiro CLI for direct tool calls -- and for sub-agent spawns from SDK 1.13.12, which on any earlier SDK is never wired as a matcher, so such a rule silently never fires. On Codex CLI and Cursor CLI that gate covers shell commands only: those are the calls the host delivers to a hook. On Kiro CLI a sub-agent's own tool calls are gated only when that sub-agent runs an agent config carrying our hooks -- one spawned without an explicit agent_name runs its own tool calls ungated once it starts, though the spawn itself is still gated, and that remaining gap is the host's. Kiro IDE governs no tool call today, and that is our choice, not a vendor gap: the installer does write the hook Kiro IDE's current loader reads, but we ship it disabled pending a soak. Kiro IDE's deterministic block also requires the v2 hook provider; on the default v1 provider a deny is mediated by the model rather than enforced. We separate those two because they mean different things to you.

See the coding agents network for full coverage.

Browser Extension

DLP for AI chat interfaces in the browser. A deny rule cancels the paste before it lands and the send before it leaves; a mask rule rewrites the matched value in place.

Shadow AI Discovery (Scout)

Planned discovery of ungoverned AI tools and endpoints operating outside your organization's governance perimeter.

Secrets Vault

Store provider keys encrypted at rest. SDK reads are policy-gated and can require human approval. Secret-read audit rows record the secret name, never the returned value.

DLP Scanning

Detect PII, secrets, and sensitive data in AI requests. Gateway response scanning is available when explicitly enabled. Includes custom regex rules and 64 built-in detectors across 12 pattern packs, including EU, Japan, Korea, Hong Kong, Singapore, Taiwan, Indonesia and Malaysia.

Policy Signing and Tamper Detection

Published policy bundles carry cryptographic signatures that SDKs verify before loading. Configure tamper handling to warn, deny one call, deny all calls, or quarantine a machine; tamper events are reported.

Air-Gap and Self-Managed

Deploy the gateway, policy engine, audit store, dashboard, and admin CLI in your infrastructure. Licensing works offline with no phone-home; policies evaluate locally.

Role-Based Access Control

Assign roles to team members with granular permissions. Control who can create policies, view audit logs, manage agents, and configure enforcement modes.

Fleet Management

Planned dashboard for enrolled devices, with status search and filtering plus per-device or fleet-wide policy-version and sync-interval configuration pushes.