ANKH Rug Pull Indicator Case Study

This repository contains a reproducible Python research pipeline for Sections 3, 4, and 5 of the project outline:

• Dataset and Data Collection
• Methodology
• Results

The pipeline analyzes public ERC-20 transfer records related to the ANKH token and associated WETH, with cautious academic framing. It is designed to work without live API access by falling back to fixture data.

Quick Start

1. Create a virtual environment with Python 3.11+.
2. Install the project:

pip install -e .[dev]

3. Optionally copy .env.example to .env and set ETHERSCAN_API_KEY.
4. Run the full fixture-backed pipeline:

make all

Commands

python scripts/01_collect_data.py --config config/config.yaml
python scripts/02_preprocess.py --config config/config.yaml
python scripts/03_run_statistics.py --config config/config.yaml
python scripts/04_run_graph_analysis.py --config config/config.yaml
python scripts/05_make_figures.py --config config/config.yaml
python scripts/06_write_results_section.py --config config/config.yaml
python scripts/07_run_qa.py --config config/config.yaml
python scripts/08_run_temporal.py --config config/config.yaml   # Phase 2: minute-level burst analysis
python scripts/09_run_community.py --config config/config.yaml  # Phase 2: Louvain community detection
python scripts/10_run_paths.py --config config/config.yaml      # Phase 2: N-hop cash-out path analysis
pytest

Notes

• Live collection uses the Etherscan V2 API when ETHERSCAN_API_KEY is available.
• If live collection is unavailable, fixture data is copied into data/raw/ and the limitation is documented in QA outputs.
• Estimated holder balances in this project are transfer-derived estimates, not guaranteed exact wallet balances.