Search the web. Read every result.

Ranked results with clean page content, so your agent reads the sources instead of guessing from links.

POST /web/search · 10 results with Markdown · 2 credits

Backed by Y Combinator. Teams build with Context.dev.

  • Mintlify
  • daily.dev
  • Daydream
  • Super.com
Credits for 10 results with Markdown
2
Median response
<1s
Results per query
100
Uptime
99.9%
Pages, not links
Each result can carry its page as Markdown, or just the passages that answer your query.
Web and news
Search the open web or recent news with the same call and the same result shape.
Filters that hold
Limit by domain, date, and country so the results match the question.

One call. Every result, readable.

Install the SDK for your language and send your first query in a few lines.

npm install context.dev

search.ts

import ContextDev from 'context.dev';const client = new ContextDev();const results = await client.web.search({  query: 'latest AI agent frameworks',  include: { markdown: true },});

Search that reads the results for you.

What it takes to get readable sources for an agent, by approach.

CompareContext.devsearch + markdownSearch APIlinks onlySearch + scrapertwo vendors
Page contentIncludedNot includedExtra request per page
Cost for 10 pages2 credits1 query1 query + 10 scrapes

Prices from each provider's public pricing. See our pricing

From your first request to a working feature.

  1. Get a key

    Sign up and copy your API key from the dashboard.

    CONTEXT_DEV_API_KEY
  2. Send a query

    Call /web/search with your query and the content you want back.

    POST /web/search
  3. Ship it

    Feed the results into your agent or your index.

A few things to know.

More details in the API documentation.

How fresh are the results?
Results come from a live search at request time.
Can I search news only?
Yes. Set the source to news to get recent articles with their dates.

Start with your next request.

One API key for scraping, research, brand data, batches, and monitoring.