Connect your data and tools to Scalt agents.
Explore the components, interfaces and deployment options available for your project.
Enterprise context.
Context is the foundation of any AI that has to act at work. Scalt’s connectors capture enterprise signals, search indexes your data, the context graph links people, customers, processes and documents, and memory learns continuously. Together they give agents everything they need to automate work, with the right permissions.
Scalt Protect.
AI needs a frame. Scalt Protect detects sensitive content before it reaches an agent, filters prompt injections and jailbreaks, anonymises personal data with GLiNER and applies the configured governance profiles.
Scalt Intelligence.
Scalt Intelligence optimises quality, cost and confidentiality on every request: each task is routed to the right model among the providers you approved, with per-team usage controls.
Open platform.
Reach Scalt from your own systems: REST API, webhooks, and external agents attached to your cases within the documented scope. Scalt fits into your environment instead of locking it in. Scalt runs as SaaS in a cloud in France, installs on your premises (on-premise) or deploys in your own cloud: AWS, Google Cloud, Azure or Scaleway.
Search, Scalter and agents rely on the same connected sources, queryable via API.
Attach your existing agents to Scalt cases, within the documented access scope.
Run agents, read cases and receive state changes in your own systems.
Governance settings include your choice of model providers and residency options, depending on the selected plan.
POST https://api.scalt.eu/v1/agents/factures-fournisseurs/runs
Authorization: Bearer SCALT_API_KEY
Content-Type: application/json
{
"input": "Traite les factures reçues cette semaine",
"profile": "finance-v3",
"approval": "inbox"
}Scalt is AI for work.
Describe the task to delegate, test the proposed agent, then share it with your team.
See agentsLet’s discuss your first use case.
Describe the work you need done and the tools you use. We’ll prepare the demonstration around your needs.
Talk with someone who knows the product and your project.
Bring a recurring task and the result you need.
We’ll explain the next steps and the conditions for beta access.