Agentic Context
Assemble and deliver relevant context to your agents on each conversational turn. The GetContext API aggregates memories, conversation history, and profile data from your data space, applies governance rules, and returns a composed context payload in an agent-friendly format optimized for LLM consumption.
REQUIRED EDITIONS
Available in: All Editions supported by Data 360. See Data 360 edition availability.
How Agentic Context Works
When an agent calls the GetContext API, Data 360 performs these operations.
- Identifies the relevant real-time data graphs for the specified individual and agent.
- Retrieves conversation data, session summaries, profile memories, and additional data from real-time data graphs.
- Applies governance filters (object-level security, field-level security, record-level security) and agent-level access control.
- Returns the context in a text format optimized for LLM consumption.
Context Sources
The GetContext API retrieves context from these real-time data graphs.
| Source Type | Data Provided |
|---|---|
| Profile | Profile-level data including extracted memories (facts, preferences) and summary information linked to the individual. |
| Conversation | Conversation-level data including recent messages, session summaries, and interaction history for the specified conversation. |
| Additional real-time data graphs | If configured during agent registration, these are included as additional data sources when available. |
Response Format
The GetContext API returns a structured response containing context entries categorized by source type. Each entry includes a source category (such as Profile or Conversation) and the context content formatted as text optimized for direct injection into an LLM prompt. The response also includes a unique context ID for tracing and debugging purposes.
[Example Pending]
For the complete response schema, field descriptions, and sample payloads, see Get Context.
Governance
Agentic Context applies governance at every stage of context assembly.
- Source data is filtered by object-level security (OLS), field-level security (FLS), and record-level security (RLS) before inclusion in context.
- Agent-level access control restricts which agents can access which data.
- All governance is applied transparently so agents receive only the data they’re authorized to see.
See Also
j