Back to Prompt Library
planning

Design LangGraph Workflow for Hybrid Search

Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.

Linked challenge: Adaptive AI Search Synthesis

Format
Text-first
Lines
1
Sections
1
Linked challenge
Adaptive AI Search Synthesis

Prompt source

Original prompt text with formatting preserved for inspection.

1 lines
1 sections
No variables
0 checklist items
Outline a LangGraph workflow that can handle a user query by first attempting an 'instant search' (e.g., direct API call). If the query is complex or ambiguous, it should transition to a 'deep thinker' path involving Gemini 2.5 Pro Deep Think for multi-step reasoning, query reformulation, and synthesis from multiple 'AI Overview' (LLM-generated summary) and traditional search results. Define the nodes, edges, and state transitions.

Adaptation plan

Keep the source stable, then change the prompt in a predictable order so the next run is easier to evaluate.

Keep stable

Preserve the role framing, objective, and reporting structure so comparison runs stay coherent.

Tune next

Swap in your own domain constraints, anomaly thresholds, and examples before you branch variants.

Verify after

Check whether the prompt asks for the right evidence, confidence signal, and escalation path.