feat: LangGraph + Memanto customer support agent with persistent cross-session memory#1
Open
crowniteto wants to merge 2 commits into
Open
feat: LangGraph + Memanto customer support agent with persistent cross-session memory#1crowniteto wants to merge 2 commits into
crowniteto wants to merge 2 commits into
Conversation
Adds a complete LangGraph integration package (integrations/langgraph/) and example (examples/langgraph-memanto/) that demonstrates: - Cross-session recall: Agent remembers findings from previous sessions - Typed semantic memory: 13 memory categories with confidence scoring - AI-driven confidence scoring: Agent self-evaluates certainty before storing - Contradiction detection: Conflicting memories flagged with versioning - Three Memanto primitives: remember, recall, answer - LangChain tool wrappers: Compatible with LangGraph prebuilt ReAct agent - Pre-built StateGraph workflow: Conditional routing based on query intent - Agent lifecycle management: MemantoSetup handles create/activate/deactivate Package: langgraph-memanto (pip-installable) Example: Customer support agent with session persistence demo Tests: 17 unit tests with mocked SdkClient (all passing) Closes moorcheh-ai#547
- Customer support agent with persistent cross-session memory via Memanto - Three-node LangGraph workflow: intake -> respond -> remember - intake node: classifies intent + recalls relevant memories from Memanto - respond node: generates LLM response enriched with recalled context - remember node: extracts and stores key insights back to Memanto - Cross-session recall: memories persist in Moorcheh cloud, not thread state - Session 2 can recall context from Session 1 with no shared state - Includes MemantoMemoryManager wrapper class - 9 unit tests covering state, nodes, JSON extraction, confidence clamping - README with architecture, quick start, and memory type documentation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a complete LangGraph integration example that demonstrates persistent cross-session memory via Memanto.
Architecture
Graph Structure
START -> intake -> respond -> remember -> ENDKey Features
Files Added
examples/langgraph-memanto/agent.py- Main agent module with MemantoMemoryManager, SupportState, graph nodes, and builderexamples/langgraph-memanto/run_example.py- Two-session demo runnerexamples/langgraph-memanto/test_agent.py- 9 unit tests (all passing)examples/langgraph-memanto/README.md- Architecture and usage documentationexamples/langgraph-memanto/requirements.txt- Python dependenciesexamples/langgraph-memanto/.env.example- Environment variable templateTest Results
Memory Types Supported
fact, preference, decision, event, observation, commitment, context, error
The agent uses the LLM to automatically extract and classify memories worth persisting across sessions.