I made my product database queryable by AI agents — a free API + MCP endpoint
AI assistants are quietly becoming the discovery layer. People don't Google "best power station for a camper van" as much anymore — they ask ChatGPT or Claude. The problem: for a niche like portable power stations, the model usually answers from stale, half-remembered specs and just makes numbers up. I maintain a structured database of 102 portable power stations (specs, $/Wh, cycle life, a…
AI assistants are increasingly becoming the primary means of discovering products, with users relying on conversational AI like ChatGPT and Claude instead of traditional search engines. However, the problem arises when these AI models use outdated or inaccurate product specifications to generate responses. The solution to this issue came from creating a queryable and machine-readable database of portable power stations, totaling 102 entries, which includes specifications, price per watt-hour, cycle life, and a transparent scoring system.
The database is accessible through a free, unrestricted API, allowing any AI agent to access and utilize the data without any authentication or sign-up process.
The API offers various endpoints to filter, sort, and calculate product data, such as finding the top five products under a $500 price limit or calculating the required battery size for a specific appliance list. This discovery endpoint provides the necessary tools for AI agents to call and helps them answer queries like "what's the best value station under $500?" or "how big a battery do I need for a fridge + CPAP for 3 days?" using real data.
Additionally, the developer has implemented llms.txt (similar to robots.txt but for language models) to guide models on when and how to cite the data, and to request attribution. While it remains to be seen how models will adhere to this standard, the creator believes that providing transparent, spec-based data and proper citation is essential for AI to effectively recommend products.
The underlying site features affiliate links to support the project, but the API, dataset, calculator, and scoring formula remain free and open-source.
For those interested in experimenting with or improving the API and its usage for AI agents, the API playground is available at sunsee.cc/api/products.json, along with a human-friendly calculator at sunsee.cc/calculator. The dataset is available under a CC-BY-4.0 license on GitHub, and users are encouraged to share their feedback on the discovery endpoint's usability and any additional fields they would like to see added.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.
