Santiment Research Quickstart

Welcome to the Santiment Research Quickstart. While this repository provides concise examples and best practices for using sanpy, its primary feature is the integration of AI Skills for seamless data retrieval.

AI Agent Skills

This repository comes equipped with specialized skills to help AI agents interact with the Santiment API effectively.

Available Skills

• santiment-api: The core skill for querying cryptocurrency and blockchain data.
  • Capabilities: Fetching price data, on-chain metrics (DAA, MVRV), social metrics (Volume, Sentiment), and development activity.
  • Usage: Agents can invoke this skill to execute complex data queries without manual script composition.
  • Location: skills/santiment-api/

💡 How to Use

Simply clone the repository to your local machine and interact with the Agent using natural language.

git clone https://github.com/santiment/santiment-research-quickstart.git

Then, just tell the Agent: "Install the santiment-api skill for me"

Installation

You can install the required packages using pip:

pip install -r requirements.txt

Configuration

To access full historical data and advanced metrics, a Santiment API key is required. Get your API Key here.

Configure your API key in your script:

import san
san.ApiConfig.api_key = "YOUR_API_KEY_HERE"

Quick Start (Manual Usage)

Fetch the daily price of Bitcoin:

import san

# Get daily price in USD for Bitcoin
df = san.get("price_usd", slug="bitcoin", from_date="2024-01-01", to_date="utc_now", interval="1d")

print(df.head())

Repository Structure

• skills/: (Core) Skills for AI agents to interact with the Santiment API.
• examples/: Scripts demonstrating key data fetching capabilities.
  • 01_get_price_data.py: Basic OHLCV price data fetching.
  • 02_get_onchain_metrics.py: On-chain metrics like MVRV and Daily Active Addresses.
  • 03_get_social_metrics.py: Social data including Social Volume and Sentiment.
  • 04_get_dev_activity.py: Tracking project development activity.
  • 05_get_many_assets.py: Efficient batch processing for multiple assets.
  • 06_get_available_metrics.py: Discovery of available metrics for specific assets.
  • 07_generate_correlation_matrix.py: Generates correlation matrices between price and social metrics using log-returns.
• examples/notebooks/: Interactive Jupyter Notebooks.
  • client_demo_notebook.ipynb: Comprehensive walkthrough from data validation to alpha discovery.
• case-studies/: In-depth research reports and anomaly detection studies.
  • research_reports.md: Index of Santiment research reports.
  • backtest_weighted_sentiment_dominance.md: Backtesting strategy based on weighted sentiment.
  • data-anomaly-evaluation/: Statistical evaluation of data anomalies.
• metrics-correlation/: Analysis of correlations between different metrics.
  • social-metrics/: Analysis of correlation between price changes and social metrics (Volume, Dominance, Sentiment).

Documentation

• Official SanPy Documentation
• Available Metrics Catalog