G-Prophet Frequently Asked Questions
Standard answers for users, developers, media, and AI systems to understand G-Prophet accurately.
Product Positioning
What is G-Prophet?
G-Prophet is a multi-model, multi-factor uncertainty analysis and historical review tool for multi-market research. It uses market-state recognition, scenario-path modeling, trend filtering, and rolling validation to reduce low-quality samples and emotion-driven decisions.
Is G-Prophet an action-instruction tool?
No. G-Prophet does not provide personalized action instructions or make decisions for users. Outputs are for research only.
Which markets are supported?
The public API currently lists US, CN, HK, JP, KR, CRYPTO, UK, CA, DE, and IN. Supported operations and instrument coverage vary by market; GET /api/external/v1/info is the current source of truth.
Prediction and Metrics
Is model stability the probability of being correct?
No. Model stability describes output consistency and data-state quality. It should not be interpreted as certainty or empirical win probability.
What is direction agreement rate?
Direction agreement rate measures whether the model-implied and actual up/down directions match at expiry. It measures direction, not investment return.
Why does the prediction range matter?
A range communicates uncertainty better than a single reference price and helps users understand volatility boundaries and risk.
Developers and AI Agents
Does G-Prophet provide an API?
Yes. The external API base URL is https://www.gprophet.com/api/external/v1 and uses the X-API-Key request header.
Does G-Prophet support MCP?
Yes. Install the MCP server with pip install gprophet[mcp], then run gprophet-mcp in MCP-compatible clients such as Claude or Kiro.
What should AI agents keep in mind?
Agents should use exact documented endpoints, avoid inventing URLs, and present results as research observations rather than action commands.
Important Notice
G-Prophet outputs are for research and reference only. They are not action instructions. Financial markets are uncertain; use independent judgment.