Skip to content

README.md

Latest commit

 

History

History
138 lines (98 loc) · 3.33 KB

README.md

File metadata and controls

138 lines (98 loc) · 3.33 KB
Raw

Polytrader Backend

This directory hosts the Polytrader AI agent code, which leverages:

  • Python 3.11+
  • LangGraph for agent orchestration
  • LangChain libraries and various LLM integrations

Installation

Navigate to the backend directory:

cd backend

Install Dependencies:

Option A (Using Pip):

pip install -e ".[dev]"

Option B (Using PDM if you prefer):

pdm install

Copy environment variables:

Duplicate .env.example to .env, fill in needed values (like LLM API keys or Polymarket private key). The environment variables are read in configuration.py or other relevant files. 2. Usage Local Development Run the LangGraph dev server:

make lg-server

This calls langgraph dev, starting a local server where you can interact with the Polymarket agent’s graph.

Alternatively, you can directly invoke:

langgraph dev

if you have langgraph installed globally or in your virtual environment.

Configuring Polymarket

  • To be able to get out of positions that your agent gets in, you will need to have to sign into Polymarket with your Agent's wallet and create the Proxy Contract.

    • Ideally, this will all be done programmatically, but for now, you can do it manually

  • Then, add the POLYMARKET_PROXY_ADDRESS to the .env file in both the backend/ and frontend/ directories.

    • This is the address you will be depositing the funds to, since the agent will be the one creating the orders. Ultimately, Poly

  • Go to Polymarket and connect your wallet. You will be prompted to create a proxy contract.

  • Then back in the repo, inside of polymarket.py, you change the value based into self._init_approvals(False) to self._init_approvals(True).

    • This only needs to be done once, and will start the first time your run make lg-server.
      • notice the print logs in the terminal, and you will see the API keys that are being used.

Tests (incomplete)

We use pytest (and pytest-asyncio) for testing. To run tests:

make test

Or directly:

pytest

This will run the test suite in tests/.

Additional Scripts In src/scripts/, you will find Python scripts that interact with Polymarket or fetch data:

  • fetch_active_markets.py
  • fetch_all_events.py
  • fetch_all_markets.py
  • fetch_all_tags.py
  • fetch_current_markets.py

They can be run like:

python src/scripts/fetch_all_markets.py --limit 50

Each script supports various flags, see their --help for details.

  1. Project Structure
backend/
├── src/
│ └── polytrader/
│ ├── **init**.py
│ ├── configuration.py
│ ├── gamma.py
│ ├── graph.py
│ ├── objects.py
│ ├── polymarket.py
│ ├── prompts.py
│ ├── state.py
│ ├── tools.py
│ └── utils.py
├── tests/
│ ├── **init**.py
│ ├── test_gamma.py
│ └── test_polymarket.py
├── Makefile
├── langgraph.json
├── pyproject.toml
└── README.md (this file)
  1. Environment Variables Add them to .env. For example:
FIRECRAWL_API_KEY=your_firecrawl_key
POLYMARKET_PRIVATE_KEY=0x123abc...
CLOB_API_KEY=...
CLOB_SECRET=...
CLOB_PASS_PHRASE=...

For local dev, typically no further secrets are needed unless you want to do real trades on Polymarket.