# @web dead? Major issue **URL:** \ **Category:** Help\ **Created:** [January 11, 2025, 5:57pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793 "2025-01-11T17:57:40Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![SoMaCoSF](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somacosf/32/14963_2.png) [@SoMaCoSF](https://forum.cursor.com/u/SoMaCoSF)\ **Post date:** [January 11, 2025, 5:57pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/1 "2025-01-11T17:57:41Z")
Why can it no lnoger @web? ![](https://us1.discourse-cdn.com/cursor1/original/3X/a/3/a36436be0687ade387b611d9ac9524a5b3610246.png) I apologize, but I don’t have direct access to web content or the ability to follow URLs directly. To help analyze the forum post and understand what the original poster (OP) is implying, could you please: - Copy and paste the relevant content from the thread, including: ---
**Author:** ![jake](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/jake/32/20459_2.png) [@jake](https://forum.cursor.com/u/jake)\ **Post date:** [January 12, 2025, 9:18am UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/2 "2025-01-12T09:18:33Z")
I believe @web means you want to search the web. Whereas if you want to index a specific URL, you should @ that URL. Example below. ![CleanShot 2025-01-12 at 17.17.59](https://us1.discourse-cdn.com/cursor1/original/3X/d/3/d355b2591a19095b651c70df49f6811707b6489a.png) ---
**Author:** ![SoMaCoSF](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somacosf/32/14963_2.png) [@SoMaCoSF](https://forum.cursor.com/u/SoMaCoSF)\ **Post date:** [January 12, 2025, 1:07pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/4 "2025-01-12T13:07:11Z")
Weird, I was doing that and it was telling me "as a bot I cant access a URL… I guess the @wrapper tells the system that its Cursor infra that will rawlthe URL as it wont pass that on to claude who cannot parse URLS… so the @ is a means to capture URLs Howeber it seems at times that can fail, somehow. ---
**Author:** ![jake](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/jake/32/20459_2.png) [@jake](https://forum.cursor.com/u/jake)\ **Post date:** [January 12, 2025, 1:46pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/5 "2025-01-12T13:46:35Z")
> [@SoMaCoSF](#): > > I guess the @wrapper tells the system that its Cursor infra that will rawlthe URL as it wont pass that on to claude who cannot parse URLS… so the @ is a means to capture URLs Yep, I think that’s right. So the raw URL without the @ confuses Claude. ---
**Author:** ![SoMaCoSF](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somacosf/32/14963_2.png) [@SoMaCoSF](https://forum.cursor.com/u/SoMaCoSF)\ **Post date:** [January 12, 2025, 2:19pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/6 "2025-01-12T14:19:52Z")
I think when it was failing, the prompt for the URL crawling was being passed to claude, rather than captured by the Cursor infra… I was looking at this yesterday: ![2025-01-11 11_34_35-Agentic Context Analysis_ Stream of Consciousness — Mozilla Firefox](https://us1.discourse-cdn.com/cursor1/original/3X/7/d/7da34bc1207b4841391537a5a21cd78c71309f44.png) ![2025-01-11 11_49_02-D__Cursor_Scratchpad_agentic_context_analysis.html](https://us1.discourse-cdn.com/cursor1/original/3X/f/9/f92415c03e92c08cf7ccda908bbbb4c744c57aa4.png) ![2025-01-11 11_50_38-D__Cursor_Scratchpad_agentic_context_analysis.html](https://us1.discourse-cdn.com/cursor1/original/3X/6/c/6c9c8a598c45e7a6083ebf805ecf86476cca68b7.png) BUt wound up fighting with Composer a bit… ---
**Author:** ![gabriel-filincowsky](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/gabriel-filincowsky/32/10043_2.png) [@gabriel-filincowsky](https://forum.cursor.com/u/gabriel-filincowsky)\ **Post date:** [January 24, 2025, 8:31pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/7 "2025-01-24T20:31:57Z")
![image](https://us1.discourse-cdn.com/cursor1/original/3X/5/3/5370a298ceef684abeef0f64747e3fd7f14c761e.png) I already tried the following: - Just “casually” adding the `@Web` in the message expecting that in the background Cursor would do its thing. - Say search the `@Web` for, then describe the issue or what I am looking for. - Say search the `@Web` for the following queries, then list the queries. - And last, add a`@Web` before each query. Note: not using VPN. ---
**Author:** ![SoMaCoSF](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somacosf/32/14963_2.png) [@SoMaCoSF](https://forum.cursor.com/u/SoMaCoSF)\ **Post date:** [January 24, 2025, 11:30pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/8 "2025-01-24T23:30:28Z")
Yeah something changed on the heavy censorship side and agentic yo has been nerfd I used to @web whole Reddit threads and have it summarize. I attempted with forums.cursor today and it gave me a super vanilla nerfd bs. Also: @cursor should have an intrinsically native agent bolted to the forum for @forum nd have it already have indexed the threads here And the agent would be lovely for: Whenever one is building. Web app, a feature to integrate an agent into your own web build such that if you’re providing a “cursor-able” forum/site/info - you can ALLOW cursors to @URL your site and get the cursor pre-formatted …cursorRSS feed. Because as we know: you shall be interviewed on the agentic workflows you bring to the table… The opposite of robots.txt - I want to feed my schema to agents so that one control then retire. Albeit site agent poisoning is the new frontier… —- Imagine @StackOverflow(thread url) already being “cursorized” so that you can feed it better. Take for example @CNCExpert (titans of cnc) they release everything for free…. Build tooling for cnc and your workflow toolchest… ---
**Author:** ![SoMaCoSF](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somacosf/32/14963_2.png) [@SoMaCoSF](https://forum.cursor.com/u/SoMaCoSF)\ **Post date:** [January 24, 2025, 11:40pm UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/9 "2025-01-24T23:40:22Z")
Yeah something changed on the heavy censorship side and agentic yo has been nerfd I used to @web whole Reddit threads and have it summarize. I attempted with forums.cursor today and it gave me a super vanilla nerfd bs. Also: @cursor should have an intrinsically native agent bolted to the forum for @forum nd have it already have indexed the threads here And the agent would be lovely for: Whenever one is building. Web app, a feature to integrate an agent into your own web build such that if you’re providing a “cursor-able” forum/site/info - you can ALLOW cursors to @URL your site and get the cursor pre-formatted …cursorRSS feed. Because as we know: you shall be interviewed on the agentic workflows you bring to the table… The opposite of robots.txt - I want to feed my schema to agents so that one control then retire. Albeit site agent poisoning is the new frontier… —- Imagine @StackOverflow(thread url) already being “cursorized” so that you can feed it better. Take for example @CNCExpert (titans of cnc) they release everything for free…. Build tooling for cnc and your workflow toolchest… ---
**Author:** ![system](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/system/32/99069_2.png) [@system](https://forum.cursor.com/u/system)\ **Post date:** [June 18, 2026, 10:25am UTC](https://forum.cursor.com/t/web-dead-major-issue/40793/10 "2026-06-18T10:25:31Z")
Closed on June 18, 2026, 10:25am UTC