Execute comprehensive Oracle research using the full autonomous agent with extended thinking.
The Oracle agent uses extended thinking capabilities to plan and execute multi-step research autonomously. It has access to a diverse set of tools including:
The agent will iteratively search and analyze your indexed repositories and documentation, perform web searches for external context, and synthesize findings into a comprehensive report.
Note: This is a long-running operation (typically 30-180 seconds). For real-time progress
updates, use the streaming endpoint /oracle/stream instead.
API key must be provided in the Authorization header
Research question to investigate
"How does authentication work in the FastAPI codebase?"
Optional list of repository identifiers to search
["fastapi/fastapi", "tiangolo/sqlmodel"]Optional list of documentation source identifiers to search
["FastAPI Documentation", "SQLModel Docs"]Optional output format specification
"markdown report with code examples"
Optional model selection for Oracle (defaults to claude-opus-4-5-20251101)
claude-opus-4-5-20251101, claude-sonnet-4-5-20250929, claude-sonnet-4-5-1m Research completed successfully
The research query that was executed
The comprehensive research report generated by Oracle
Tool-result citations recorded during the run
List of tool calls made during research
Number of research iterations performed
Duration of the research in milliseconds
Unique session identifier for this research
When this research session was created