🧩 LangMiddle — Composable Middlewares for LangGraph

Lightweight, plug‑and‑play middlewares for LangGraph developers.

Overview

Production-ready middleware for LangChain and LangGraph v1 with multi-backend chat history persistence. Store conversations in SQLite, Supabase, or Firebase with zero configuration required.

Key Features:

• ✅ LangChain / LangGraph v1 Compatible: Native middleware pattern support
• ⚡ Zero Config Start: Defaults to in-memory SQLite—no setup needed
• 🔄 Multi-Backend Storage: Switch between SQLite, PostgreSQL, Supabase, Firebase with one parameter
• 🔒 Production Ready: JWT authentication, RLS support, type-safe
• 🧠 Context Engineering: Built-in support for long-term facts, semantic memories, and context management

Available middleware

Middleware	Description	Supported Backends
ToolRemover	Removes tool-related messages from the conversation state (pre/post agent).	N/A (no backend needed)
ChatSaver	Persists chat histories	SQLite ✅, PostgreSQL ✅, Supabase ✅, Firebase ✅
ContextEngineer	Auto context management with semantic memories	Supabase 🚧 (in progress)

Installation

Core Package (SQLite only):

pip install langmiddle

With Optional Backends:

# For PostgreSQL support
pip install langmiddle[postgres]

# For Supabase support
pip install langmiddle[supabase]

# For Firebase support
pip install langmiddle[firebase]

# All backends
pip install langmiddle[all]

Quick Start - LangChain Middleware

from langmiddle.history import ChatSaver, ToolRemover, StorageContext

# Initialize middleware with desired backend
agent = create_agent(
    model="gpt-4o",
    tools=[],
    context_schema=StorageContext,
    middleware=[
        ToolRemover(),
        ChatSaver(backend="sqlite", db_path="./chat_history.db")
    ],
)

agent.invoke(
    input={"messages": [{"role": "user", "content": "Hello!"}]},
    context={
        "user_id": "***",           # User ID in UUID format
        "session_id": "***",        # Thread ID in UUID format
    },
)

ContextEngineer

ContextEngineer provides automatic context and memory management for conversations. It can store, retrieve, and query long-term facts (semantic memories) using the project's supported storage backends.

from langchain.chat_models import init_chat_model
from langchain.embeddings import init_embeddings
from langmiddle.context import ContextEngineer

# Initialize model and embedder
llm = init_chat_model("gpt-4o")
embedder = init_embeddings("text-embedding-3-small")

# When running for **the first time**, create databases in the backend
# Requires connection string for table creations
store = ChatStorage.create(
    "supabase",
    auto_create_tables=True,
    enable_facts=True,
)

# Use in your agent
agent = create_agent(
    model="gpt-4o",
    tools=[],
    context_schema=StorageContext,
    middleware=[
        ContextEngineer(
            model=llm,
            embedder=embedder,
            backend="supabase",
            backend_kwargs={'enable_facts': True},
        ),
    ],
)

agent.invoke(
    input={"messages": [{"role": "user", "content": "Tell me about my preferences."}]},
    context={
        "user_id": "***",           # User ID in UUID format
        "session_id": "***",        # Thread ID in UUID format
        "auth_token": "***",        # JWT token for supabase backend
    },
)

Storage Backends

Backend	Use Case	Pros	Cons	Setup
SQLite	Development, Single-user	Simple, Local, Fast, No setup	Not distributed	None
PostgreSQL	Self-hosted, Custom auth	Full control, Standard SQL, Flexible	Manual setup	Connection string
Supabase	Production Web Apps	Managed PostgreSQL, RLS, Real-time	Supabase-specific	Environment vars
Firebase	Mobile, Google ecosystem	Real-time, Managed, Global	Google-specific	Service account

SQLite Configuration

# Local file
backend_type="sqlite", db_path="./chat.db"

# In-memory (testing)
backend_type="sqlite", db_path=":memory:"

PostgreSQL Configuration

# .env file or environment variables
POSTGRES_CONNECTION_STRING=postgresql://user:password@localhost:5432/dbname

# Or pass directly
backend_type="postgres",
connection_string="postgresql://user:password@localhost:5432/dbname",
auto_create_tables=True

See PostgreSQL Setup Guide for details.

Supabase Configuration

# .env file or environment variables
SUPABASE_URL=your_supabase_url
SUPABASE_ANON_KEY=your_anon_key

# To create tables
SUPABASE_CONNECTION_STRING=your_connection_string

Firebase Configuration

# Service account credentials file
backend_type="firebase", credentials_path="./firebase-creds.json"

# Or use GOOGLE_APPLICATION_CREDENTIALS environment variable

---

Contributing

We welcome contributions! If LangMiddle helped you, please ⭐️ the repo to help others discover it.

License

Apache License 2.0 - see LICENSE for details.