Welcome to Skymith Technical Documentation¶

Getting Started¶

The best way to see how Skymith works is to try it! You can create an account for free in https://skymith.com. Afterwards, you will be assigned API key that you will need throughout this tutorial.

Holmes¶

Holmes is our service for searching the web automatically, so that you can build fast your own research agents in any topic. It has two modes of operation:

• Speed: focused on retrieving the information fast and accurately, not in depth.
• Quality: focused on providing complete answers, with great depth and breath. Really useful for open-ended questions and/or really specific research.

Strider¶

Strider is our service for comparing products across different platforms that you provide. The information to retrieve for the product is defined in the query itself to improve the flexibility of the product.

Important: you must have permission to retrieve information from those platforms, and our user agent Mitheithel must be allowed in their robots.txt for this service to work with them.

Buffet: Public Financial Information¶

To power our service GPTStonks (https://gptstonks.net), we have developed a set of tools that we now serve as services:

• We research and summarize the most important financial news of the day, everyday. With our Daily Financial News endpoint you can have access to our summaries.
• We propose a financial concept to learn every day.

Data Policies¶

We never keep your data, and neither do the third-party services we use. All your requests are treated in real-time and any intermediate artifact that could be generated is removed afterwards.

Contributing 🤝¶

We welcome contributions from the community! Feel free to open issues, report bugs, or propose changes if you want to contribute to the project.

Disclaimer¶

Skymith serves as an interface for accessing all kinds of data and general knowledge. It is not intended for unethical or illegal purposes. Always check your local legislation before using our services.

We always respect robots.txt, so if you don't want your web to be included in our results just exclude our agent "Mitheithel".