One API to scrape, enrich, and understand the web. Trusted by 20K+ developers.
- Tweets 848
- Following 326
- Followers 3,110
- Likes 644
Context.dev retweeted
You’re solving my use cases faster than I can try using Context for them
1
1
5
1,263
Context.dev retweeted
PSA: new case study unlocked 🔓
@sazabi is building a worldclass AI-native observability platform
When a customer signs up, they automatically discover their tech stack and recommend how to instrument it.
But scanning a codebase only shares so much. Analytics, support widgets, and other scripts often only show up after JavaScript runs.
So they added Context.dev to scan the website too :)
> rendered HTML reveals client-side tools the codebase scan misses
> each detection points to a script, asset host, or response header
> signup to live in production in minutes
The whole discovery pipeline is deterministic, with our API supplying the rendered pages.
Awesome seeing @shcallaway turn web data into something immediately useful for his customers :)
5
3
38
1,606
Context.dev retweeted
Love this marketing idea!! Free coffee just outside of yc
@inkbox_ai @getcontextdev
12
2
68
4,966
Context.dev retweeted
Just tried the @getcontextdev API and yeah, I'm really impressed. I told my agent about it, gave it the docs URL, Codex immediately showed me sample code, gave the OK, and I was up and running in under 2 minutes. This is a keeper.
1
1
3
1,995
Introducing Context Highlights: web extraction for agents 97% more token-efficient than Codex.
Given a URL and a query, it extracts relevant passages with surrounding context to help the agent interpret them.
Agents using naive fetch and scrape tools are polluting their context and multiplying costs.
Compaction helps, but it's better to keep irrelevant text out of the context window in the first place.
17
12
93
45,937
We also compared Highlights with Zilliz's open-source semantic highlighting model, giving both the same source text from SimpleQA Verified.
We were able to match its answer accuracy with 3x fewer tokens.
1
1
4
751
On Exa's WebCode benchmark, we matched Exa Highlights' groundedness with about 20% fewer output tokens.
Against Claude Code's WebFetch, we preserved more of the evidence at almost the same token count: 76% groundedness versus 50.4%.
Claude's WebFetch has a webpage to summarize pages and leave out relevant details. In contrast, Context Highlights is purely extractive.
Read our research: context.dev/blog/introducing…
Add Highlights to an agent you're building:
docs.context.dev/scrape/high…
Add Highlights to your Claude Code or Codex agent: MCP setup:
docs.context.dev/install-mcp
1
3
416
Scrape anything.
Introducing Scrape Products
Every store has a different website, your agent shouldn't care
> input: a product URL
> output: prices, images, variants, availability, descriptions, brands & more, all as clean JSON
No custom parsers for every store, digging through HTML, or babysitting CSS selectors.
Build shopping agents, price trackers, product comparisons, and more :)
3
8
1,329
Context.dev retweeted
Building an AI agent for web scraping, research, lead generation, or data extraction?
Introducing @swytchcode for Scraping 🚀
You can now connect your agents to scraping APIs like @apify , @AgentQL , hystruck, @scrapegraphai , @browserless , axesso, @getcontextdev
3
4
10
829
The highest quality web data API for the best price
Now with benefits for startups
Introducing context.dev for startups
30% off any plan, for a full year.
Apply here context.dev/startups
2
9
1,295
i heard they're pretty good
1
14
1,140
Let's go!
shipped: real screenshots inside articles
when @seoladders writes an article it already goes and captures the actual pages it references.
a screenshot is evidence. an AI-generated image is decoration.
but sometimes it misses one.
so now you can just ask. "screenshot the semrush pricing page and put it in section 2"
the agent captures it, shows you where it wants to put it, and waits.
you approve. it never edits your article on its own.
capture is powered by @getcontextdev.
genuinely impressed so far, it's what the agent uses to fetch from the web.
8
954
Context.dev retweeted
This launch is built on top of tools built by incredible companies including
@unkeydev @getcontextdev @paper @eve/@aisdk (by @vercel) @neondatabase @autumnpricing @WorkOS @hoplite_sh @typesafeai @EffectTS_ and many many more
Without them this launch wouldn't have been possible. I highly recommend trying these out!
Does ChatGPT know you exist?
Someone asks it for the best tool in your category. It names two competitors. You never hear about it.
This happens more and more every month.
Notra GEO finds where AI engines skip you, then writes the content to fix it.
Be the answer.
9
4
43
2,042
Context.dev retweeted
Introducing Scrape Anything.
we merged all our scraping APIs into one
inputs: websites, PDFs, DOCX, PPT, XLSX, 40 file types, anything.
into HTML, markdown, screenshots, images, JSON & more
Give us a URL, get your data
No need to worry about bot blocking, js rendering, proxies, iframes, shadow doms, retries, etc..
All the weird edge cases are our problem :)
104
72
1,679
230,951
short but good list
i get 30 pitches a day on linkedin and email here is
saas i actually bought and use
1. @getcontextdev
2. @Fiber_AI
3. @ahrefs
4. @WisprFlow
2
2,022
We're offering grants of up to $20,000 in Context credits to researchers working on RL, agents, and evals.
Academic labs, startups, open-source teams, and independent researchers are very welcome.
Some applications we're especially excited about:
-building jev-like models on web data
-improving general model capability with better web data
-improving design taste/ui with our screenshot api
-benchmarking search/scraping APIs
-fast searchers
Apply at context.dev/research
3
3
29
3,175
