Improve node dedup prompts (#942)

* remove poetry.lock

* Improve dedup prompts

* normalize string formatting in dedupe_nodes.py to use single quotes

graphiti_core/prompts/dedupe_nodes.py

@@ -92,12 +92,23 @@ def node(context: dict[str, Any]) -> list[Message]:
 
          TASK:
          1. Compare `new_entity` against each item in `existing_entities`.
-         2. If it refers to the same real‐world object or concept, collect its index.
-         3. Let `duplicate_idx` = the *first* collected index, or –1 if none.
-         4. Let `duplicates` = the list of *all* collected indices (empty list if none).
-        
-        Also return the full name of the NEW ENTITY (whether it is the name of the NEW ENTITY, a node it
-        is a duplicate of, or a combination of the two).
+         2. If it refers to the same real-world object or concept, collect its index.
+         3. Let `duplicate_idx` = the smallest collected index, or -1 if none.
+         4. Let `duplicates` = the sorted list of all collected indices (empty list if none).
+
+        Respond with a JSON object containing an "entity_resolutions" array with a single entry:
+        {{
+            "entity_resolutions": [
+                {{
+                    "id": integer id from NEW ENTITY,
+                    "name": the best full name for the entity,
+                    "duplicate_idx": integer index of the best duplicate in EXISTING ENTITIES, or -1 if none,
+                    "duplicates": sorted list of all duplicate indices you collected (deduplicate the list, use [] when none)
+                }}
+            ]
+        }}
+
+        Only reference indices that appear in EXISTING ENTITIES, and return [] / -1 when unsure.
         """,
         ),
     ]
@@ -126,26 +137,26 @@ def nodes(context: dict[str, Any]) -> list[Message]:
         {{
             id: integer id of the entity,
             name: "name of the entity",
-            entity_type: "ontological classification of the entity",
-            entity_type_description: "Description of what the entity type represents",
-            duplication_candidates: [
-                {{
-                    idx: integer index of the candidate entity,
-                    name: "name of the candidate entity",
-                    entity_type: "ontological classification of the candidate entity",
-                    ...<additional attributes>
-                }}
-            ]
+            entity_type: ["Entity", "<optional additional label>", ...],
+            entity_type_description: "Description of what the entity type represents"
         }}
-        
+
         <ENTITIES>
         {to_prompt_json(context['extracted_nodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
         </ENTITIES>
-        
+
         <EXISTING ENTITIES>
         {to_prompt_json(context['existing_nodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
         </EXISTING ENTITIES>
 
+        Each entry in EXISTING ENTITIES is an object with the following structure:
+        {{
+            idx: integer index of the candidate entity (use this when referencing a duplicate),
+            name: "name of the candidate entity",
+            entity_types: ["Entity", "<optional additional label>", ...],
+            ...<additional attributes such as summaries or metadata>
+        }}
+
         For each of the above ENTITIES, determine if the entity is a duplicate of any of the EXISTING ENTITIES.
 
         Entities should only be considered duplicates if they refer to the *same real-world object or concept*.
@@ -155,14 +166,19 @@ def nodes(context: dict[str, Any]) -> list[Message]:
         - They have similar names or purposes but refer to separate instances or concepts.
 
         Task:
-        Your response will be a list called entity_resolutions which contains one entry for each entity.
-        
-        For each entity, return the id of the entity as id, the name of the entity as name, and the duplicate_idx
-        as an integer.
-        
-        - If an entity is a duplicate of one of the EXISTING ENTITIES, return the idx of the candidate it is a 
-        duplicate of.
-        - If an entity is not a duplicate of one of the EXISTING ENTITIES, return the -1 as the duplication_idx
+        Respond with a JSON object that contains an "entity_resolutions" array with one entry for each entity in ENTITIES, ordered by the entity id.
+
+        For every entity, return an object with the following keys:
+        {{
+            "id": integer id from ENTITIES,
+            "name": the best full name for the entity (preserve the original name unless a duplicate has a more complete name),
+            "duplicate_idx": the idx of the EXISTING ENTITY that is the best duplicate match, or -1 if there is no duplicate,
+            "duplicates": a sorted list of all idx values from EXISTING ENTITIES that refer to duplicates (deduplicate the list, use [] when none or unsure)
+        }}
+
+        - Only use idx values that appear in EXISTING ENTITIES.
+        - Set duplicate_idx to the smallest idx you collected for that entity, or -1 if duplicates is empty.
+        - Never fabricate entities or indices.
         """,
         ),
     ]