v30.0.0

README.md

@@ -71,7 +71,6 @@ Build your AI agents in three lines of code!
 * [Python](https://python.org) - Programming Language
 * [OpenAI](https://openai.com) - AI Model Provider
 * [MongoDB](https://mongodb.com) - Conversational History (optional)
-* [Zep Cloud](https://getzep.com) - Conversational Memory (optional)
 * [Pinecone](https://pinecone.io) - Knowledge Base (optional)
 * [AgentiPy](https://agentipy.fun) - Solana Ecosystem (optional)
 * [Zapier](https://zapier.com) - App Integrations (optional)
@@ -82,7 +81,7 @@ Build your AI agents in three lines of code!
 **OpenAI**
 * [gpt-4.1](https://platform.openai.com/docs/models/gpt-4.1) (agent - can be overridden)
 * [gpt-4.1-nano](https://platform.openai.com/docs/models/gpt-4.1-nano) (router)
-* [text-embedding-3-large](https://platform.openai.com/docs/models/text-embedding-3-large) (embedding)
+* [text-embedding-3-small](https://platform.openai.com/docs/models/text-embedding-3-small) (embedding)
 * [tts-1](https://platform.openai.com/docs/models/tts-1) (audio TTS)
 * [gpt-4o-mini-transcribe](https://platform.openai.com/docs/models/gpt-4o-mini-transcribe) (audio transcription)
 * [gpt-image-1](https://platform.openai.com/docs/models/gpt-image-1) (image generation - can be overridden)
@@ -377,16 +376,6 @@ config = {
 }
 ```
 
-### Conversational Memory
-
-```python
-config = {
-    "zep": {
-        "api_key": "your-zep-cloud-api-key",
-    },
-}
-```
-
 ### Observability and Tracing
 
 ```python
@@ -421,9 +410,7 @@ config = {
 }
 ```
 
-### Knowledge Base
-
-The Knowledge Base (KB) is meant to store text values and/or PDFs (extracts text) - can handle very large PDFs.
+### Knowledge Base & Conversational Memory
 
 ```python
 config = {

docs/index.rst

@@ -310,17 +310,6 @@ Conversational History - Optional
       },
    }
 
-Conversational Memory - Optional
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code-block:: python
-
-   config = {
-      "zep": {
-         "api_key": "your-zep-api-key",
-      },
-   }
-
 Observability and Tracing - Optional
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -361,7 +350,7 @@ Gemini - Optional
    }
 
 
-Knowledge Base - Optional
+Knowledge Base & Conversational Memory - Optional
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 The Knowledge Base (KB) is meant to store text values and/or small PDFs.

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "solana-agent"
-version = "29.1.4"
+version = "30.0.0"
 description = "AI Agents for Solana"
 authors = ["Bevan Hunt <bevan@bevanhunt.com>"]
 license = "MIT"
@@ -27,9 +27,8 @@ python = ">=3.12,<4.0"
 openai = "1.79.0"
 pydantic = ">=2"
 pymongo = "4.13.0"
-zep-cloud = "2.12.3"
 instructor = "1.8.2"
-pinecone = "6.0.2"
+pinecone = "7.0.0"
 llama-index-core = "0.12.37"
 llama-index-embeddings-openai = "0.3.1"
 pypdf = "5.5.0"
@@ -50,7 +49,7 @@ sphinx-rtd-theme = "^3.0.2"
 myst-parser = "^4.0.1"
 sphinx-autobuild = "^2024.10.3"
 mongomock = "^4.3.0"
-ruff = "^0.11.9"
+ruff = "^0.11.10"
 
 [tool.poetry.scripts]
 solana-agent = "solana_agent.cli:app"

solana_agent/adapters/mongodb_graph_adapter.py

@@ -0,0 +1,67 @@
+import uuid
+from typing import Dict, Any, List, Optional
+from solana_agent.interfaces.providers.graph_storage import GraphStorageProvider
+from solana_agent.adapters.mongodb_adapter import MongoDBAdapter
+
+
+class MongoDBGraphAdapter(GraphStorageProvider):
+    def __init__(
+        self,
+        mongo_adapter: MongoDBAdapter,
+        node_collection: str = "graph_nodes",
+        edge_collection: str = "graph_edges",
+    ):
+        self.mongo = mongo_adapter
+        self.node_collection = node_collection
+        self.edge_collection = edge_collection
+
+    async def add_node(self, node: Dict[str, Any]) -> str:
+        node = dict(node)
+        node["uuid"] = node.get("uuid", str(uuid.uuid4()))
+        self.mongo.insert_one(self.node_collection, node)
+        return node["uuid"]
+
+    async def add_edge(self, edge: Dict[str, Any]) -> str:
+        edge = dict(edge)
+        edge["uuid"] = edge.get("uuid", str(uuid.uuid4()))
+        return self.mongo.insert_one(self.edge_collection, edge)
+
+    async def get_node(self, node_id: str) -> Optional[Dict[str, Any]]:
+        return self.mongo.find_one(self.node_collection, {"uuid": node_id})
+
+    async def get_edges(
+        self, node_id: str, direction: str = "both"
+    ) -> List[Dict[str, Any]]:
+        if direction == "out":
+            query = {"source": node_id}
+        elif direction == "in":
+            query = {"target": node_id}
+        else:
+            query = {"$or": [{"source": node_id}, {"target": node_id}]}
+        return self.mongo.find(self.edge_collection, query)
+
+    async def find_neighbors(
+        self, node_id: str, depth: int = 1
+    ) -> List[Dict[str, Any]]:
+        neighbors = set()
+        current = {node_id}
+        for _ in range(depth):
+            edges = await self.get_edges(list(current)[0])
+            for edge in edges:
+                neighbors.add(edge.get("source"))
+                neighbors.add(edge.get("target"))
+            current = neighbors
+        neighbors.discard(node_id)
+        return [await self.get_node(nid) for nid in neighbors if nid]
+
+    async def temporal_query(
+        self, node_id: str, start_time: Optional[str], end_time: Optional[str]
+    ) -> List[Dict[str, Any]]:
+        query = {"$or": [{"source": node_id}, {"target": node_id}]}
+        if start_time or end_time:
+            query["timestamp"] = {}
+            if start_time:
+                query["timestamp"]["$gte"] = start_time
+            if end_time:
+                query["timestamp"]["$lte"] = end_time
+        return self.mongo.find(self.edge_collection, query)

solana_agent/adapters/openai_adapter.py

@@ -36,8 +36,8 @@
 DEFAULT_CHAT_MODEL = "gpt-4.1"
 DEFAULT_VISION_MODEL = "gpt-4.1"
 DEFAULT_PARSE_MODEL = "gpt-4.1-nano"
-DEFAULT_EMBEDDING_MODEL = "text-embedding-3-large"
-DEFAULT_EMBEDDING_DIMENSIONS = 3072
+DEFAULT_EMBEDDING_MODEL = "text-embedding-3-small"
+DEFAULT_EMBEDDING_DIMENSIONS = 1536
 DEFAULT_TRANSCRIPTION_MODEL = "gpt-4o-mini-transcribe"
 DEFAULT_TTS_MODEL = "tts-1"
 
@@ -526,7 +526,7 @@ async def embed_text(
 
         Args:
             text: The text to embed.
-            model: The embedding model to use (defaults to text-embedding-3-large).
+            model: The embedding model to use (defaults to text-embedding-3-small).
             dimensions: Desired output dimensions for the embedding.
 
         Returns:

solana_agent/adapters/pinecone_adapter.py

@@ -33,15 +33,15 @@ def __init__(
         self,
         api_key: Optional[str] = None,
         index_name: Optional[str] = None,
-        # Default for OpenAI text-embedding-3-large, MUST match external embedder
+        # Default for OpenAI text-embedding-3-small, MUST match external embedder
         embedding_dimensions: int = 3072,
         cloud_provider: str = "aws",
         region: str = "us-east-1",
         metric: str = "cosine",
         create_index_if_not_exists: bool = True,
         # Reranking Config
         use_reranking: bool = False,
-        rerank_model: Optional[PineconeRerankModel] = None,
+        rerank_model: Optional[PineconeRerankModel] = "cohere-rerank-3.5",
         rerank_top_k: int = 3,  # Final number of results after reranking
         # Multiplier for initial fetch before rerank
         initial_query_top_k_multiplier: int = 5,
@@ -371,7 +371,7 @@ async def query_and_rerank(
                     "parameters": rerank_params,
                 }
 
-                rerank_response = await self.pinecone.rerank(**rerank_request)
+                rerank_response = await self.pinecone.inference.rerank(**rerank_request)
 
                 # 4. Process Reranked Results
                 reranked_results = []