What does “good” mean for us?
Generic benchmarks don’t reflect your standards or your risk appetite.
Leitwerk tests your AI agents against your own standards, proves the result with signed evidence, and tells you when an approval no longer holds.
The Readiness Check is a free 45-minute conversation.
Three questions block every agent release.
Generic benchmarks don’t reflect your standards or your risk appetite.
Manual reviews are slow and leave no reproducible evidence.
Models, prompts and tools change. Yesterday’s test may not hold today.
How it works
What you get
One dashboard shows which agents are approved, against which standards, and until when.
Built for audit. Anyone can verify it on a second machine: signature, traces, scores and verdict are recomputed.
Less manual review, and a decision that is the same every time you run it.
On premise
EU-hosted or in your own environment. On premise, Leitwerk runs fully inside your walls and needs no outbound connection.
CPU-only containers. No GPU needed.
Outbound connections needed. It runs fully on premise.
Your agent talks to the gate, not to your real ERP, CRM or mail.
Vertex AI EU, Azure OpenAI EU, STACKIT, local vLLM or Ollama
ADK, LangGraph, OpenAI Agents SDK or any MCP-capable framework
Design Partner Program 2027
A pilot on one real agent workflow, with the first companies that want approvals they can defend.
It starts with one page that names four things.
AI Agent Readiness Check
A free conversation about where your agents stand. We score six areas from 1 to 5. Within two working days you get a scorecard and three recommendations.
We also look at what makes it urgent:
We will get back to you to find a time. If anything is urgent, write to hello@leitwerk.tech.