AI agents that work securely with your data.
Our platforms are built to work with AI: dedicated assistants inside the application, and an open API and MCP that let your own AI agents work with them securely.
Assistants inside the application
In Activaflow, assistant Iris answers questions about the dossier, the software and the law, with sources. AI proposes, the user decides.
Open API & MCP
Through the API and the Model Context Protocol, external tools and AI agents, such as Claude, can work with Activaflow.
Per-key permissions
Each key can only do what you allow, and everything goes into the audit log. That way, what an agent has done always remains verifiable.
Agentic AI in-house
We also use AI agents internally, for example to handle email. What we put to the test in practice ourselves, we bring to our platforms.
Frequently asked questions
Can I connect my own AI assistant?
With Activaflow you can, via the API and MCP. Each key can only do what you allow.
How do you prevent an agent from being able to do too much?
With per-key permissions and an audit log. An agent can only do what you explicitly allow.
How our platforms are built.
- 01 Models We do not use AI for the sake of AI. Self-trained models, lightning-fast decision models, generative imagery and language models: applied where they measurably take work off your hands.
- 02 Workflows Automation only works if you can trust it. That is why we build in approval steps, audit logs and locks.
- 03 Integrations Our platforms work with the software you already use. And if something is not supported out of the box, we build the integration.
- 04 Infrastructure Fast software on a solid foundation. Hosted in Europe, with backups we actually restore to test them.
Built on our own terms.
webRidge has been self-built since 2011, without loans or external investors. We own our platforms and choose the long term: not absurd margins, but lasting relationships with our customers.