Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.langchain.com/llms.txt

Use this file to discover all available pages before exploring further.

Privy is powerful wallet infrastructure for AI agents, built for scale.

​
Overview

Create agents that can:
  • Automatically create and manage wallets
  • Make payments in a variety of digital assets, including stablecoins
  • Sign messages and transactions
  • Query wallet balances and addresses

​
How it works

Privy provides wallet infrastructure that removes the complexity of blockchain interactions:
  1. Give your agent a powerful wallet
  2. Secure your agent’s assets with transaction policies
  3. Make payments with speed
Zero-friction onboarding Different from traditional wallet solutions, Privy automatically creates embedded wallets for your agents without requiring private key management, seed phrases, or complex setup. Production-ready infrastructure Trusted by leading Web3 applications, Privy handles secure key generation, multi-chain address derivation, transaction signing, and compliance policies at scale.

​
Quickstart

import os
from langchain_privy import PrivyWalletTool
from langchain.agents import create_agent

# Set credentials
os.environ["PRIVY_APP_ID"] = "your-privy-app-id"
os.environ["PRIVY_APP_SECRET"] = "your-privy-app-secret"

# Initialize wallet tool (automatically creates wallet)
privy_tool = PrivyWalletTool()
print(f"Wallet created! Address: {privy_tool.wallet_address}")

# Create agent
agent = create_agent(
    model="claude-sonnet-4-6",
    tools=[privy_tool],
)

# Agent can now perform wallet operations
agent.invoke({"messages": [{"role": "user", "content": "What's my wallet address on Base?"}]})
See the complete example for a full implementation.

​
Setup

Head to Privy’s dashboard to sign up and create a new app. You’ll receive:
  • App ID - Your application identifier
  • App Secret - Your server-side authentication secret
  1. Install the package:
pip install langchain-privy
  1. Set your credentials:
import os
import getpass

os.environ["PRIVY_APP_ID"] = getpass.getpass("Enter your Privy App ID: ")
os.environ["PRIVY_APP_SECRET"] = getpass.getpass("Enter your Privy App Secret: ")

​
Instantiation

from langchain_privy import PrivyWalletTool

# Automatically creates a new Ethereum wallet
tool = PrivyWalletTool()

# Or create on a specific chain
base_tool = PrivyWalletTool(chain_type="base")
solana_tool = PrivyWalletTool(chain_type="solana")

# Or reuse an existing wallet
existing_tool = PrivyWalletTool(wallet_id="wal_abc123...")

​
Invocation

​
Available operations

# Get wallet address for any chain
tool.invoke({
    "operation": "get_wallet_address",
    "chain": "base"
})

# Sign a message
tool.invoke({
    "operation": "sign_message",
    "message": "Hello from LangChain!",
    "chain": "ethereum"
})

# Check balance
tool.invoke({
    "operation": "get_balance",
    "chain": "base"
})

# Send transaction
tool.invoke({
    "operation": "send_transaction",
    "chain": "base",
    "to": "0x1234567890123456789012345678901234567890",
    "value": "0.001",
    "unit": "ether"
})

​
Use within an agent

import os
from langchain_privy import PrivyWalletTool
from langchain.agents import create_agent

# Set credentials
os.environ["PRIVY_APP_ID"] = "your-privy-app-id"
os.environ["PRIVY_APP_SECRET"] = "your-privy-app-secret"

# Initialize tools
tools = [PrivyWalletTool()]

# Create agent
agent = create_agent(
    model="claude-sonnet-4-6",
    tools=tools,
)

# Natural language wallet operations
agent.invoke({
    "messages": [{"role": "user", "content": "Sign the message 'Verified by AI Agent' and then check my balance on Base"}]
})

​
API reference

Connect these docs to Claude, VSCode, and more via MCP for real-time answers.
Edit this page on GitHub or file an issue.