Connect your data estate to any AI model, without writing a line of integration code.
Intelligence implements the Model Context Protocol, the emerging open standard for connecting AI agents to enterprise tools. That means every model that speaks MCP can reach every Symphony extension through a single connection, with no vendor-specific glue code in between.
Works with
Claude · ChatGPT · watsonx.ai Agent Lab · Mistral · local models (Ollama, LM Studio, Open WebUI) · any MCP host
Reaches
Canon · Ice Flow · Iceberg Ops · Pulse · Observability · Data Migrator · any Cirata extension with an OpenAPI spec
Let watsonx.ai, Claude, or ChatGPT call your replication, lakehouse, and monitoring extensions directly, without bespoke wrappers per model.
Serve Claude on the developer laptop, ChatGPT for ad-hoc analysis, watsonx.ai in production, and a local model behind the firewall, all from one catalogue.
Answer the regulator's question — which agent did what, to which data, on whose behalf, with tenant-scoped sessions, validated arguments, and end-to-end traces.
Meter AI usage as a first-class platform dimension. Bill by client, by query, by catalogue size, through the same chargeback your data services already use.
Add one flag to your OpenAPI spec.
Intelligence finds the tool, parses the schema, publishes it to every connected AI host.
The model calls. Intelligence validates, routes, traces, and meters every request.
