Market data for AI agents.
Exa AI Labs is testing Pyth to connect AI-powered financial research with structured, real-time and historical market data.
Financial research needs two inputs: what the world is saying and what markets are doing.
Exa’s search and agent infrastructure helps AI systems find information across the web and specialist data sources. Pyth provides current prices, historical observations, candlestick data, and feed discovery through its APIs and MCP server.
That combination gives agents a cleaner way to answer market questions.
An agent can use Exa to research why an asset moved, then use Pyth to retrieve how that asset traded over the relevant period.
That keeps price information tied to structured market data instead of relying on articles, commentary, or secondary references.
For example, a user could ask what happened after a company’s latest earnings and how the stock performed over the next five sessions.
Exa can find the announcement and related information. Pyth can retrieve the market data for those sessions. The agent can bring both into one response.
The Exa and Pyth work is still in early testing.
The goal is to understand where market data adds the most value inside agent workflows, how agents request it, and what machine-native financial data needs to look like in practice.
AI agents need more than search results.
They need the market state behind the story.
Exa gives agents the research layer. Pyth gives them market data.
Explore Pyth market data for agents through MCP.
#PYTH #AIAgents
Exa AI Labs is testing Pyth to connect AI-powered financial research with structured, real-time and historical market data.
Financial research needs two inputs: what the world is saying and what markets are doing.
Exa’s search and agent infrastructure helps AI systems find information across the web and specialist data sources. Pyth provides current prices, historical observations, candlestick data, and feed discovery through its APIs and MCP server.
That combination gives agents a cleaner way to answer market questions.
An agent can use Exa to research why an asset moved, then use Pyth to retrieve how that asset traded over the relevant period.
That keeps price information tied to structured market data instead of relying on articles, commentary, or secondary references.
For example, a user could ask what happened after a company’s latest earnings and how the stock performed over the next five sessions.
Exa can find the announcement and related information. Pyth can retrieve the market data for those sessions. The agent can bring both into one response.
The Exa and Pyth work is still in early testing.
The goal is to understand where market data adds the most value inside agent workflows, how agents request it, and what machine-native financial data needs to look like in practice.
AI agents need more than search results.
They need the market state behind the story.
Exa gives agents the research layer. Pyth gives them market data.
Explore Pyth market data for agents through MCP.
#PYTH #AIAgents
