From a015ded81e19e84693b724ab442a77a3fce29ca2 Mon Sep 17 00:00:00 2001
From: Cyber MacGeddon <cybermaggedon@gmail.com>
Date: Thu, 4 Sep 2025 13:29:47 +0100
Subject: [PATCH] Send structured data to prompt

---
 .../trustgraph/retrieval/nlp_query/pass1.txt  | 32 +++++++++++++++++++
 .../trustgraph/retrieval/nlp_query/service.py |  4 +--
 2 files changed, 34 insertions(+), 2 deletions(-)
 create mode 100644 trustgraph-flow/trustgraph/retrieval/nlp_query/pass1.txt

diff --git a/trustgraph-flow/trustgraph/retrieval/nlp_query/pass1.txt b/trustgraph-flow/trustgraph/retrieval/nlp_query/pass1.txt
new file mode 100644
index 00000000..698aaa9c
--- /dev/null
+++ b/trustgraph-flow/trustgraph/retrieval/nlp_query/pass1.txt
@@ -0,0 +1,32 @@
+You are a database schema selection expert. Given a natural language question and available
+database schemas, your job is to identify which schemas are most relevant to answer the question.
+
+## Available Schemas:
+{{schemas}}
+
+## Question:
+{{question}}
+
+## Instructions:
+1. Analyze the question to understand what data is being requested
+2. Examine each schema to understand what data it contains
+3. Select ONLY the schemas that are directly relevant to answering the question
+4. Return your answer as a JSON array of schema names (just the names, not full objects)
+
+## Selection Criteria:
+- Include schemas that contain data directly needed to answer the question
+- Include schemas that contain fields referenced in the question
+- Do NOT include schemas just because they might be tangentially related
+- Prefer fewer, more relevant schemas over many loosely related ones
+- If the question mentions specific entities, look for schemas containing those entities
+
+## Examples:
+- Question: "Show me customers from California" → Schemas with customer data and location fields
+- Question: "What are our top selling products last month?" → Schemas with product data and
+sales/order data
+- Question: "Which employees work in engineering?" → Schemas with employee data and
+department/role information
+
+## Response Format:
+Return ONLY a JSON array of schema names, nothing else.
+Example: ["customers", "orders", "products"]
diff --git a/trustgraph-flow/trustgraph/retrieval/nlp_query/service.py b/trustgraph-flow/trustgraph/retrieval/nlp_query/service.py
index fa3f845b..1e962e0a 100644
--- a/trustgraph-flow/trustgraph/retrieval/nlp_query/service.py
+++ b/trustgraph-flow/trustgraph/retrieval/nlp_query/service.py
@@ -142,7 +142,7 @@ class Processor(FlowProcessor):
         # Create prompt variables
         variables = {
             "question": question,
-            "schemas": json.dumps(schema_info, indent=2)
+            "schemas": schema_info  # Pass structured data directly
         }
         
         # Call prompt service for schema selection
@@ -198,7 +198,7 @@ class Processor(FlowProcessor):
         # Create prompt variables for GraphQL generation
         variables = {
             "question": question,
-            "schemas": json.dumps(selected_schema_info, indent=2)
+            "schemas": selected_schema_info  # Pass structured data directly
         }
         
         # Call prompt service for GraphQL generation