GetOutpost MCP Server
A Model Context Protocol (MCP) server that brings real-time Indian options market data and volatility analytics from GetOutpost.in into your AI chats.
What is this?
This MCP server enables Claude to analyze options market data including:
Implied Volatility (IV) - Options pricing and market expectations
Realized Volatility (RV) - Historical price movement metrics (Close-to-Close, Parkinson, Garman-Klass, Rogers-Satchell, Yang-Zhang)
Volatility Risk Premium (VRP) - IV divided by RV
Skew Analysis - Volatility smile/smirk patterns across strike prices
Perfect for quantitative analysis, options trading research, and data-driven trading insights on Indian markets (NSE, BSE).
Installation
Option 1: Claude Desktop Extension Marketplace (Coming Soon)
Status: Submitted and under review by Anthropic
Once approved, you'll be able to install directly from the marketplace:
Open Claude Desktop
Go to Settings → Extensions
Search for "GetOutpost Financial Data"
Click Install
Configure your credentials file path when prompted
Option 2: Direct Download (.mcpb file)
Available Now! Download and install the extension bundle directly:
Download the latest release: getoutpost-mcp-server-v1.0.0.mcpb
Double-click the downloaded
.mcpbfile, orDrag and drop it into Claude Desktop, or
In Claude Desktop, go to Settings → Extensions → Install Extension and select the file
Configure your credentials file path when prompted
Option 3: NPM Global Installation
Install globally via npm:
Then configure:
Getting GetOutpost Credentials
Sign up at GetOutpost.in
Log in and obtain your access token and refresh token from dev tools.
Create a credentials file at
~/.getoutpost_credentials.json:
Note: The MCP server automatically refreshes your tokens and updates this file when tokens expire.
Available Tools
The server provides 8 tools that Claude can use:
Data Retrieval Tools
Tool | Description |
| Get implied volatility data for specific symbols |
| Get realized volatility data using various calculation methods |
| Get volatility risk premium (IV minus RV) |
| Get volatility skew patterns across strikes |
Discovery/Filtering Tools
Tool | Description |
| Find symbols by IV percentile range |
| Find symbols by RV percentile range |
| Find symbols by VRP percentile range |
| Find symbols by skew percentile range |
Available Prompts
Pre-built prompt templates to help you get started:
Find long volatility opportunities - Discover symbols where buying options makes sense (default: 30 DTE)
Find cheap OTM puts - Locate extremely cheap deep out-of-the-money puts for tail hedging (default: 29 DTE)
Find short volatility opportunities - Identify symbols for selling options strategies (default: 30 DTE)
Find optimal short vol conditions - Advanced filtering for moderately priced IV with healthy RV to avoid mean reversion (default: 25 DTE, ATM)
Access these via the Prompts menu in Claude Desktop.
Example Usage
Once installed, you can ask Claude questions like:
Or use the pre-built prompts for guided workflows.
Recommended Claude Setup
For best results:
Create a Claude Project named "Options Insights"
Project Description: "Access real-time options market data and volatility analytics through GetOutpost's financial APIs. Analyze implied volatility, realized volatility, volatility risk premium, and skew across multiple instruments to generate data-driven trading insights."
Add Custom Instructions: Copy the content from
system_prompt.mdand paste it into your project's custom instructions. This provides Claude with optimal guidance on how to use the tools effectively.Use Claude Sonnet 4 for optimal analysis
Start fresh chats when switching between different symbols or analysis types to avoid context length issues
Understanding Key Parameters
Moneyness:
log(Forward/Strike)- Positive values = OTM puts, 0 = ATM, Negative = OTM callsDays to Expiry (DTE): Calendar days until option expiration
Volatility Types:
c2c- Close-to-Closeparkinson- Parkinson's range-based estimatorgarman_klass- Garman-Klass estimatorrogers_satchell- Rogers-Satchell estimatoryang_zhang- Yang-Zhang estimatormean- Average of all methods
Lookback Period: Days of historical data (20, 40, 60, or 80)
Token Management
The server handles authentication automatically:
Detects when your access token expires (401 error)
Uses your refresh token to obtain new credentials
Updates your credentials file with the new tokens
Retries the failed request seamlessly
You never need to manually refresh tokens.
Development
Building from Source
Running Tests
Project Structure
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
MIT License - see LICENSE file for details.
Support
Issues: GitHub Issues
GetOutpost Support: GetOutpost.in
MCP Documentation: Model Context Protocol
Acknowledgments
Built with the Model Context Protocol SDK by Anthropic.
Disclaimer: This tool is for informational and educational purposes only. Always conduct your own research and consult with financial advisors before making trading decisions.