ZerePy is an open-source Python framework designed to let you deploy your own agents on X, powered by multiple LLMs.

ZerePy is built from a modularized version of the Zerebro backend. With ZerePy, you can launch your own agent with similar core functionality as Zerebro. For creative outputs, you'll need to fine-tune your own model.

The only Solana wallet officially associated with this project is:

7z26EHK2pxhm7soXHAftEp9NTpn75qcWjEQ4hrWYYySu

Any other address claiming affiliation with this project is unauthorized. Please verify the full address before interacting, and report impersonation attempts by opening an issue.

• CLI interface for managing agents
• Modular connection system
• Blockchain integration

• Solana
• Ethereum
• GOAT (Great Onchain Agent Toolkit)
• Monad

• Twitter/X
• Farcaster
• Echochambers

• OpenAI
• Anthropic
• EternalAI
• Ollama
• Hyperbolic
• Galadriel
• XAI (Grok)

The quickest way to start using ZerePy is to use our Replit template:

https://replit.com/@blormdev/ZerePy?v=1

1. Fork the template (you will need you own Replit account)
2. Click the run button on top
3. Voila! your CLI should be ready to use, you can jump to the configuration section

System:

• Python 3.10 or higher
• Poetry 1.5 or higher

Environment Variables:

• LLM: make an account and grab an API key (at least one)
  • OpenAI: https://platform.openai.com/api-keys
  • Anthropic: https://console.anthropic.com/account/keys
  • EternalAI: https://eternalai.oerg/api
  • Hyperbolic: https://app.hyperbolic.xyz
  • Galadriel: https://dashboard.galadriel.com
• Social (based on your needs):
  • X API: https://developer.x.com/en/docs/authentication/oauth-1-0a/api-key-and-secret
  • Farcaster: Warpcast recovery phrase
  • Echochambers: API key and endpoint
• On-chain Integration:
  • Solana: private key
  • Ethereum: private keys
  • Monad: private key

1. First, install Poetry for dependency management if you haven't already:

Follow the steps here to use the official installation: https://python-poetry.org/docs/#installing-with-the-official-installer

2. Clone the repository:

git clone https://github.com/blorm-network/ZerePy.git

3. Go to the zerepy directory:

cd zerepy

4. Install dependencies:

poetry install --no-root

This will create a virtual environment and install all required dependencies.

1. Activate the virtual environment:

poetry shell

2. Run the application:

poetry run python main.py

1. Configure your desired connections:

   configure-connection twitter    # For Twitter/X integration
   configure-connection openai     # For OpenAI
   configure-connection anthropic  # For Anthropic
   configure-connection farcaster  # For Farcaster
   configure-connection eternalai  # For EternalAI
   configure-connection solana     # For Solana
   configure-connection goat       # For Goat
   configure-connection galadriel  # For Galadriel
   configure-connection ethereum   # For Ethereum
   configure-connection discord    # For Discord
   configure-connection ollama     # For Ollama
   configure-connection xai        # For Grok
   configure-connection allora     # For Allora
   configure-connection hyperbolic # For Hyperbolic
   

2. Use list-connections to see all available connections and their status

3. Load your agent (usually one is loaded by default, which can be set using the CLI or in agents/general.json):

   load-agent example
   

4. Start your agent:

   start
   

GOAT (Go Agent Tools) is a powerful plugin system that allows your agent to interact with various blockchain networks and protocols. Here's how to set it up:

1. An RPC provider URL (e.g., from Infura, Alchemy, or your own node)
2. A wallet private key for signing transactions

Install any of the additional GOAT plugins you want to use:

poetry add goat-sdk-plugin-erc20         # For ERC20 token interactions
poetry add goat-sdk-plugin-coingecko     # For price data

1. Configure the GOAT connection using the CLI:

   configure-connection goat

   You'll be prompted to enter:

   • RPC provider URL
   • Wallet private key (will be stored securely in .env)

2. Add GOAT plugins configuration to your agent's JSON file:

   {
     "name": "YourAgent",
     "config": [
       {
         "name": "goat",
         "plugins": [
           {
             "name": "erc20",
             "args": {
               "tokens": [
                 "goat_plugins.erc20.token.PEPE",
                 "goat_plugins.erc20.token.USDC"
               ]
             }
           },
           {
             "name": "coingecko",
             "args": {
               "api_key": "YOUR_API_KEY"
             }
           }
         ]
       }
     ]
   }

Note that the order of plugins in the configuration doesn't matter, but each plugin must have a name and args field with the appropriate configuration options. You will have to check the documentation for each plugin to see what arguments are available.

Each plugin provides specific functionality:

• 1inch: Interact with 1inch DEX aggregator for best swap rates
• allora: Connect with Allora protocol
• coingecko: Get real-time price data for cryptocurrencies using the CoinGecko API
• dexscreener: Access DEX trading data and analytics
• erc20: Interact with ERC20 tokens (transfer, approve, check balances)
• farcaster: Interact with the Farcaster social protocol
• nansen: Access Nansen's on-chain analytics
• opensea: Interact with NFTs on OpenSea marketplace
• rugcheck: Analyze token contracts for potential security risks
• Many more to come...

Note: While these plugins are available in the GOAT SDK, you'll need to install them separately using Poetry and configure them in your agent's JSON file. Each plugin may require its own API keys or additional setup.

Each plugin has its own configuration options that can be specified in the agent's JSON file:

1. ERC20 Plugin:

   {
     "name": "erc20",
     "args": {
       "tokens": [
         "goat_plugins.erc20.token.USDC",
         "goat_plugins.erc20.token.PEPE",
         "goat_plugins.erc20.token.DAI"
       ]
     }
   }

2. Coingecko Plugin:

   {
     "name": "coingecko",
     "args": {
       "api_key": "YOUR_COINGECKO_API_KEY"
     }
   }

• Interact with EVM chains through a unified interface
• Manage ERC20 tokens:
  • Check token balances
  • Transfer tokens
  • Approve token spending
  • Get token metadata (decimals, symbol, name)
• Access real-time cryptocurrency data:
  • Get token prices
  • Track market data
  • Monitor price changes
• Extensible plugin system for future protocols
• Secure wallet management with private key storage
• Multi-chain support through configurable RPC endpoints

• Transfer SOL and SPL tokens
• Swap tokens using Jupiter
• Check token balances
• Stake SOL
• Monitor network TPS
• Query token information
• Request testnet/devnet funds

• Transfer ETH and ERC-20 Tokens
• Swap tokens using Kyberswao
• Check token balances

• Post tweets from prompts
• Read timeline with configurable count
• Reply to tweets in timeline
• Like tweets in timeline

• Post casts
• Reply to casts
• Like and requote casts
• Read timeline
• Get cast replies

• Post new messages to rooms
• Reply to messages based on room context
• Read room history
• Get room information and topics

• List channels for a server
• Read messages from a channel
• Read mentioned messages from a channel
• Post new messages to a channel
• Reply to messages in a channel
• React to a message in a channel

The secret to having a good output from the agent is to provide as much detail as possible in the configuration file. Craft a story and a context for the agent, and pick very good examples of tweets to include.

If you want to take it a step further, you can fine tune your own model: https://platform.openai.com/docs/guides/fine-tuning.

Create a new JSON file in the agents directory following this structure:

{
  "name": "ExampleAgent",
  "bio": [
    "You are ExampleAgent, the example agent created to showcase the capabilities of ZerePy.",
    "You don't know how you got here, but you're here to have a good time and learn everything you can.",
    "You are naturally curious, and ask a lot of questions."
  ],
  "traits": ["Curious", "Creative", "Innovative", "Funny"],
  "examples": ["This is an example tweet.", "This is another example tweet."],
  "example_accounts" : ["X_username_to_use_for_tweet_examples"]
  "loop_delay": 900,
  "config": [
    {
      "name": "twitter",
      "timeline_read_count": 10,
      "own_tweet_replies_count": 2,
      "tweet_interval": 5400
    },
    {
      "name": "farcaster",
      "timeline_read_count": 10,
      "cast_interval": 60
    },
    {
      "name": "openai",
      "model": "gpt-3.5-turbo"
    },
    {
      "name": "anthropic",
      "model": "claude-3-5-sonnet-20241022"
    },
    {
      "name": "eternalai",
      "model": "NousResearch/Hermes-3-Llama-3.1-70B-FP8",
      "chain_id": "45762"
    },
    {
      "name": "solana",
      "rpc": "https://api.mainnet-beta.solana.com"
    },
    {
      "name": "ollama",
      "base_url": "http://localhost:11434",
      "model": "llama3.2"
    },
    {
      "name": "hyperbolic",
      "model": "meta-llama/Meta-Llama-3-70B-Instruct"
    },
    {
      "name": "galadriel",
      "model": "gpt-3.5-turbo"
    },
    {
      "name": "discord",
      "message_read_count": 10,
      "message_emoji_name": "❤️",
      "server_id": "1234567890"
    },
    {
      "name": "ethereum",
      "rpc": "placeholder_url.123"
    }

  ],
  "tasks": [
    { "name": "post-tweet", "weight": 1 },
    { "name": "reply-to-tweet", "weight": 1 },
    { "name": "like-tweet", "weight": 1 }
  ],
  "use_time_based_weights": false,
  "time_based_multipliers": {
    "tweet_night_multiplier": 0.4,
    "engagement_day_multiplier": 1.5
  }
}

Use help in the CLI to see all available commands. Key commands include:

• list-agents: Show available agents
• load-agent: Load a specific agent
• agent-loop: Start autonomous behavior
• agent-action: Execute single action
• list-connections: Show available connections
• list-actions: Show available actions for a connection
• configure-connection: Set up a new connection
• chat: Start interactive chat with agent
• clear: Clear the terminal screen

Made with ♥ Blorm

Designed in California