implementing chroma

swarms/memory/base_vector_db.py

@@ -18,13 +18,13 @@ class VectorDatabase(ABC):
 
     @abstractmethod
     def query(
-        self, vector: Dict[str, Any], num_results: int
+        self, text: str, num_results: int
     ) -> Dict[str, Any]:
         """
         Query the database for vectors similar to the given vector.
 
         Args:
-            vector (Dict[str, Any]): The vector to compare against.
+            text (Dict[str, Any]): The vector to compare against.
             num_results (int): The number of similar vectors to return.
 
         Returns:

swarms/memory/chroma.py

@@ -1,10 +1,13 @@
-from typing import List
+from abc import ABC, abstractmethod
+from typing import Any, Dict, List
 from chromadb.utils import embedding_functions
 from httpx import RequestError
 import chromadb
 
+from swarms.memory.base_vector_db import VectorDatabase
 
-class ChromaClient:
+
+class ChromaClient(VectorDatabase):
     def __init__(
         self,
         collection_name: str = "chromadb-collection",
@@ -90,23 +93,14 @@ class ChromaClient:
             print(f"Error searching vectors: {e}")
             return None
 
-    def search_vectors_formatted(self, query: str, limit: int = 2):
-        """
-        Searches the collection for vectors similar to the query vector.
+    def add(self, vector: Dict[str, Any], metadata: Dict[str, Any]) -> None:
+        pass
 
-        Args:
-            query (str): The query string to be converted into a vector and used for searching.
-            limit (int): The number of search results to return. Defaults to 3.
+    def query(self, vector: Dict[str, Any], num_results: int) -> Dict[str, Any]:
+        pass
 
-        Returns:
-            SearchResult or None: Returns the search results if successful, otherwise None.
-        """
-        try:
-            search_result = self.collection.query(
-                                    query_texts=query,
-                                    n_results=limit,
-                                )
-            return search_result
-        except Exception as e:
-            print(f"Error searching vectors: {e}")
-            return None
+    def delete(self, vector_id: str) -> None:
+        pass
+
+    def update(self, vector_id: str, vector: Dict[str, Any], metadata: Dict[str, Any]) -> None:
+        pass

swarms/structs/agent.py

@@ -635,6 +635,10 @@ class Agent:
                     AGENT_SYSTEM_PROMPT_3, response
                 )
 
+                # Retreiving long term memory
+                if self.memory:
+                    task = self.agent_memory_prompt(response, task)
+
                 attempt = 0
                 while attempt < self.retry_attempts:
                     try:
@@ -755,6 +759,33 @@ class Agent:
             """
             return agent_history_prompt
 
+    def agent_memory_prompt(
+            self,
+            query,
+            prompt
+    ):
+        """
+        Generate the agent long term memory prompt
+
+        Args:
+            system_prompt (str): The system prompt
+            history (List[str]): The history of the conversation
+
+        Returns:
+            str: The agent history prompt
+        """
+        context_injected_prompt = prompt
+        if self.memory:
+            ltr = self.memory.query(query)
+
+            context_injected_prompt = f"""{prompt}
+                ################ CONTEXT ####################
+                {ltr}
+            """
+
+        return context_injected_prompt
+
+
     async def run_concurrent(self, tasks: List[str], **kwargs):
         """
         Run a batch of tasks concurrently and handle an infinite level of task inputs.