feat: composio integration — agent memory primitive + tool contracts + behavior protocol

- Add synapse_memory/integrations/langchain_memory.py (SynapseChatMessageHistory)
- Modern BaseChatMessageHistory adapter (langchain-core >=0.3)
- Full sync + async support
- Messages routed through Cognitive Security pipeline
- Add examples/langchain_agent.py with LCEL pattern
- Add tests/integrations/test_langchain_memory.py (22 tests)
- Add langchain optional dependency in pyproject.toml
- Update README with LangChain integration section
- 287 tests passing, 94% coverage, zero IP exposure

Author: Ismael Marchi

README.md

@@ -212,6 +212,46 @@ engine = AutoSaveEngine(
 
 ---
 
+## 🔗 Using Synapse Layer with LangChain
+
+Synapse Layer integrates natively with LangChain as a chat message history backend.
+Every message passes through the full Cognitive Security pipeline — PII redaction,
+intent validation, and AES-256 encryption — before persistence.
+
+```bash
+pip install synapse-layer langchain-core
+```
+
+```python
+from synapse_memory.integrations import SynapseChatMessageHistory
+
+# Drop-in replacement for any LangChain message history
+history = SynapseChatMessageHistory(agent_id="my-agent")
+
+history.add_user_message("I prefer concise responses.")
+history.add_ai_message("Got it — keeping it brief.")
+
+# Retrieve messages in LangChain-compatible format
+messages = history.messages
+```
+
+Works with `RunnableWithMessageHistory` for LCEL chains:
+
+```python
+from langchain_core.runnables.history import RunnableWithMessageHistory
+
+chain_with_history = RunnableWithMessageHistory(
+    runnable=your_chain,
+    get_session_history=lambda sid: SynapseChatMessageHistory(
+        agent_id="your-agent", session_id=sid,
+    ),
+)
+```
+
+> **Note:** This is the OSS adapter. Advanced scoring, enterprise retrieval
+> strategies, and PRO heuristics are available under separate license.
+> See [synapselayer.org/docs](https://synapselayer.org/docs).
+
 ## Competitive Comparison
 
 | Capability | Synapse Layer | Mem0 | Zep | pgvector (raw) |
@@ -265,7 +305,7 @@ Synapse Layer is not just a library you call — it's **infrastructure your agen
 | Version | Status | Highlights |
 |---|---|---|
 | **v1.0.7** | ✅ **Stable** | Auto-Save Engine, Plugin Architecture, MCP Bridge, Smithery listing |
-| **v1.1.0** | 🚧 In Progress | LangChain native adapter, CrewAI integration, embedding model selection |
+| **v1.1.0** | 🚧 In Progress | ✅ LangChain native adapter, CrewAI integration, embedding model selection |
 | **v1.2.0** | 📋 Planned | Synapse Forge visual debugger, real-time memory inspector |
 | **v2.0.0** | 📋 Planned | Multi-tenant vault, team memory spaces, RBAC |
 

examples/langchain_agent.py

@@ -0,0 +1,66 @@
+"""
+Synapse Layer — LangChain Integration Example
+
+Demonstrates how to use Synapse Layer as persistent memory
+for a LangChain conversational agent.
+
+Requirements:
+    pip install synapse-layer langchain-core
+
+Author : Security & Architecture Team @ Synapse Layer
+License: Apache 2.0
+"""
+
+import asyncio
+from synapse_memory.integrations import SynapseChatMessageHistory
+
+
+async def main():
+    # Initialize Synapse Layer as your LangChain message history
+    history = SynapseChatMessageHistory(
+        agent_id="demo-agent",
+        session_id="onboarding-session",
+    )
+
+    # Store conversation messages
+    # Each message passes through the Cognitive Security pipeline:
+    # PII redaction → Intent validation → AES-256 encryption
+    history.add_user_message("I prefer concise, technical responses.")
+    history.add_ai_message("Noted — I'll keep responses brief and precise.")
+    history.add_user_message("My project deadline is next Friday.")
+    history.add_ai_message("I'll factor that deadline into my suggestions.")
+
+    print("Stored 4 messages through Synapse Layer.")
+    print(f"Memories in vault: {len(history._memory._memories)}")
+    print()
+
+    # Retrieve messages in LangChain-compatible format
+    messages = await history.aget_messages()
+    print(f"Retrieved {len(messages)} messages:")
+    for msg in messages:
+        role = "User" if msg.type == "human" else "AI"
+        print(f"  [{role}] {msg.content}")
+    print()
+
+    # Use with RunnableWithMessageHistory (LangChain LCEL):
+    #
+    # from langchain_core.runnables.history import RunnableWithMessageHistory
+    #
+    # chain_with_history = RunnableWithMessageHistory(
+    #     runnable=your_chain,
+    #     get_session_history=lambda session_id: SynapseChatMessageHistory(
+    #         agent_id="your-agent",
+    #         session_id=session_id,
+    #     ),
+    # )
+    #
+    # response = chain_with_history.invoke(
+    #     {"input": "What were my preferences?"},
+    #     config={"configurable": {"session_id": "user-123"}},
+    # )
+
+    print("Done. Memory persists across sessions with zero-knowledge security.")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

pyproject.toml

@@ -45,6 +45,9 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
+langchain = [
+  "langchain-core>=0.3.0",
+]
 dev = [
   "pytest>=7.4.0",
   "pytest-cov>=4.1.0",
@@ -64,7 +67,7 @@ Changelog = "https://github.com/SynapseLayer/synapse-layer/blob/main/CHANGELOG.m
 Security = "https://github.com/SynapseLayer/synapse-layer/blob/main/SECURITY.md"
 
 [tool.setuptools]
-packages = ["synapse_memory", "synapse_memory.engine", "synapse_memory.crypto"]
+packages = ["synapse_memory", "synapse_memory.engine", "synapse_memory.crypto", "synapse_memory.autosave", "synapse_memory.plugins", "synapse_memory.integrations"]
 
 [tool.setuptools.package-dir]
 "" = "."

synapse_memory/__init__.py

@@ -50,6 +50,8 @@
 __all__ = [
     # Core
     "SynapseMemory",
+    # Integrations (lazy — import from synapse_memory.integrations)
+    # "SynapseChatMessageHistory",
     "StoreResult",
     "RecallResult",
     # Sanitizer

synapse_memory/integrations/__init__.py

@@ -0,0 +1,12 @@
+"""
+Synapse Layer — Ecosystem Integrations
+
+Official adapters for popular AI frameworks.
+
+Author : Security & Architecture Team @ Synapse Layer
+License: Apache 2.0
+"""
+
+from .langchain_memory import SynapseChatMessageHistory
+
+__all__ = ["SynapseChatMessageHistory"]