Give your coding agent current context

Stop your agent guessing from stale training data: connect live docs, search and scraping.

You end up with: Your agent answers library questions from current docs and can pull any web page you name into context.

1. Context7

Up-to-date library documentation and code examples for LLMs, straight from the source.

Why: Pulls version-correct documentation for the exact libraries in your project; the fastest fix for outdated-API answers.

2. Exa

Neural web search for agents — find pages by meaning, not keywords.

Why: Meaning-based web search when the question is broader than one library — error messages, comparisons, current events.

3. Firecrawl Optional

Turn websites into clean markdown for your agent — crawl, scrape and search.

Why: Turns any specific page you already know about into clean markdown the agent can read.

Setup instructions and compatibility are compiled from publisher docs and community reports, not hands-on testing, unless a step is marked Tested. Check the source before relying on a setup.