AI, Agents & Machine Interfaces

Expose your application to AI models through MCP, accept delegated work from other agents through A2A, publish an OpenAPI specification, and keep building with the AI Builder.

• MCP Server — Let an AI model use your application as a tool source, with credentials and a capability ladder controlling exactly what it may do.
• A2A Server — Accept delegated tasks from other AI agents, advertise what you are genuinely good at, and delegate outward to agents you have allowlisted.
• OpenAPI Specification — Generate a machine-readable API specification from the datastores you have chosen to expose, and decide what belongs on that surface.
• AI Builder & Large Uploads — Keep changing your application by describing what you want, including from spreadsheets you upload, long after the first version was built.
• AI Providers & Settings — Set up the AI providers and keys, choose which model each AI feature uses, and configure Gemini and Amazon Bedrock from the AI Settings screen.
• AI Usage & Run Log — See what AI costs and how often it fails, test a role with a live call, control what the AI run log keeps, and run AI work in the background.
• AI Credit & Spend Limits — Charge AI use to tenants from a plan allowance and top-ups, read a tenant's AI Credit screen, and report AI spend to ActiveManage under an installation cap.
• AI Images & Video — Generate images and short videos on Google Gemini: who may use it, daily allowances, the models offered, and how videos are collected and paid for.
• TypeSafe Jev — Typed AI decisions from TypeSafe Jev: switching it on, the console and playground, AI workflow steps, Is about (AI) message rules and Feed Queue suggestions.