| What it does | Execution governance for consequential AI-agent operations
| Open-source runtime for computer-use agents
|
|---|
| Stage | Not given
| Startup
|
|---|
| Founded | Not given
| 2026
|
|---|
| Based in | Not given
| Not given
|
|---|
| Pricing model | Freemium
| Open source
|
|---|
| Pricing | Free engine (MIT/Apache-2.0 open source). Team tier: €1,900/month including 25 writing seats and 50,000 governed actions per month; overage €0.02 per action above 50k monthly. Enterprise: from €90,000/year. Design partner program: €18,000 for 90 days (credited against first annual contract). Add-ons
| Not given
|
|---|
| Key features | - Authority revalidation at dispatch time against live conditions
- Approval binding to canonical intent with single-use, time-bound enforcement
- Sequence and prerequisite policy enforcement at plan time and runtime
- Effect identity matching to prevent duplicate actions across runs
- HMAC-signed, hash-chained, and Ed25519-anchored evidence
- Cross-session capacity ledger for aggregate exposure control
- Verifiers in five languages for independent offline verification
- MCP server support for integration with Claude Code, Cursor, and other agents
- Air-gapped deployment with documented key ceremony on Enterprise
- Deterministic Rust enforcement engine with no models in the decision path
| - Works with any OpenAI-compatible model
- Ready-wait mechanism instead of fixed sleeps
- Batches actions and bounds model calls
- Desktop pause, resume and fork with memory
- MCP server for Claude Code and Cursor
- Graded rollouts for computer-use RL
- Works with full desktops
- Supports multiple sandbox providers
|
|---|
| What makes it different | - Authority is revalidated at the moment of execution, not just at approval time
- Signed, verifiable evidence allows third-party audit without trusting the vendor
- Deterministic enforcement with no AI models influencing the gate decision
- Pricing on governed actions, not seats, aligned with consequence not headcount
- Refusals and verification are never metered or charged
| - 1.48x faster than Anthropic's computer-use reference loop on the same model
- 19 points more success than OSWorld's Qwen agent at same speed
- Reduces rollout time by 40-53% by eliminating unnecessary waiting
- Published benchmarks run same model on same desktops with reproducible results
- Works with any model and sandbox provider
|
| Community votes | 0 | 0 |
| DR (Domain Rating by Ahrefs) | 1 ▼ −1 (down) since last check | Not checked yet ● no change since last check |
| Trust Flow (Majestic) | 0 | 0 |
| Citation Flow (Majestic) | 0 | 0 |
| Referring domains (Majestic) | 0 | 0 |