frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Ask HN: Agents Craving Context

1•AnthonyR•1h ago
Hey, I was an early adopter of the Cursor IDE and pretty much installed it day 1 of it's public release. Like many I've been using LLMs to help accelerate the writing of code in my own company and side projects and have been quite optimistic and satisfied with the quality and experience of using the Cursor IDE chat -> composer -> agents to enhance the developer experience.

However, recently it seems like there has been a shift in how agents work (I am mainly referring to using Cursor agents here but the seems to also apply for Claude Code). When I introduce a problem to the agent a while ago it used to just start writing code. I remember having to be explicit about selecting the files or even lines I wanted it to edit or generate code in, and be quite precise in supplying the LLM with the relevant context. These days I will write a similar prompt to what I would do before: "Refactor X.ts into separate files /utils/[module].ts using the /shared/utils directory as an example". (just a toy example)

Now, instead of just reading the examples I feed it and the context I provide, the agent will read the files and notice the imports, it reads into which imports may be important and then goes into the node_modules and reads the library code and then realizes there are docs linked and then it asks to fetch the docs linked there then in the docs there is a reference to another library so it needs to fetch that library which links to a Github page which it needs to read before realiing.. wait what was the task again? Oh yeah refactoring this. But wait now I need to read all the files that rely on this to make sure that I can refactor those correcly once I make these changes... and so on.

This was not the case even just 1 year ago. Before I was able to do a clear job defining the task and the LLM would do the task with the provided information. Sometimes I would provide the wrong information and it would do the task wrong and I could undo, fix the information and have to do the task again, or I could manually go in a fix the issues. Now, the agents are consuming huge amounts of context and taking incredibly large amounts of time to produce "correct" code where in my experience the 1-shot accuracy seems to only have marginally improved due to the fact that when they finally get to the step of generating code there is so much extraneous context in their prompting that can cause artifacts or issues in the actual generated solution, not to mention the fact that a simple request will take maybe a full minute to change some code in 1 file...

Anyways apologies for the rant but I was wondering if anyone else is experiencing similar changes to the AI coding space and if they have come up with solutions for it? (Caveman mode maybe ? :))

NIST cuts down CVE analysis amid vulnerability overload

https://www.csoonline.com/article/4159882/nist-cuts-down-cve-analysis-amid-vulnerability-overload...
1•WaitWaitWha•55s ago•0 comments

Our Tax System Should Make You Furious

https://www.nytimes.com/2026/04/17/opinion/ezra-klein-podcast-ray-madoff.html
1•keernan•4m ago•0 comments

Claude Opus 4.7 Dropped and My Trust Got a Little Smaller

https://www.bhusalmanish.com.np/blog/posts/opus-4-7-developer-fatigue.html
1•okchildhood•4m ago•0 comments

When AI agents show up to class

https://moodle.com/news/field-notes-when-ai-agents-show-up-to-class/
1•the-mitr•5m ago•0 comments

Suspects used smart glasses, other AI tools to commit 'organized' retail fraud

https://www.ctvnews.ca/toronto/article/suspects-used-smart-glasses-other-ai-tools-to-commit-organ...
1•KLK2019•5m ago•1 comments

Norway Launches First Driverless Bus in Regular Traffic

https://www.nrk.no/rogaland/historisk_-kolumbus-kjorer-uten-sjafor-pa-bussen-i-stavanger-1.17841909
1•chaosprint•7m ago•0 comments

Claude Opus 4.7 costs 20–30% more per session

https://www.claudecodecamp.com/p/i-measured-claude-4-7-s-new-tokenizer-here-s-what-it-costs-you
2•aray07•7m ago•0 comments

Average new UK electric car price is now lower than petrol vehicles

https://www.theguardian.com/environment/2026/apr/17/new-uk-electric-car-price-petrol-ev-autotrader
5•MBCook•10m ago•1 comments

From Vivaldi to Van Halen, classical and heavy metal are a natural pairing

https://www.theguardian.com/music/2026/apr/15/from-strads-to-shreds-and-vivaldi-to-van-halen-clas...
2•robtherobber•11m ago•0 comments

Show HN: Amendments to the PayPal Cryptocurrency Terms and Conditions

https://www.paypal.com/us/legalhub/paypal/upcoming-policies-full
1•ppolicyco•11m ago•0 comments

What if the robots came for the org chart instead?

https://substack.com/home/post/p-194053020
1•sirnicolaz•11m ago•0 comments

Saturn's largest moon could see 10-foot waves from a tiny breeze

https://www.popsci.com/science/saturn-moon-titan-big-waves/
2•Brajeshwar•12m ago•0 comments

A new direction for AI developer tooling featuring creator of Tidewave

https://changelog.com/friends/112
1•fagnerbrack•13m ago•0 comments

Tips for Probabilistic Software

https://jxnl.co/writing/2024/01/19/tips-probabilistic-software/
1•fagnerbrack•13m ago•0 comments

Netflix plans to add a vertical video feed

https://techcrunch.com/2026/04/17/netflix-plans-to-add-a-vertical-video-feed-use-ai-for-recommend...
1•Levitating•15m ago•0 comments

The Gregorio project – GPL tools for typesetting Gregorian chant

https://gregorio-project.github.io/index.html
3•mcookly•16m ago•0 comments

The tide has turned in the AI infrastructure space

https://radiant.co/blog/the-tide-is-receding-in-ai-infrastructure
1•theaiguyhack•17m ago•0 comments

App Store Reviews Are Busted

https://blog.terrygodier.com/2026/04/13/app-store-reviews-are-busted.html
1•gpi•18m ago•0 comments

Kamal

https://github.com/basecamp/kamal
3•sakopov•18m ago•0 comments

Agents, Code Reviews, and the Bottleneck Shift, Oh My

https://ayende.com/blog/203939-C/agents-code-reviews-and-the-bottleneck-shift-oh-my?Key=a4aac51e-...
1•speckx•18m ago•0 comments

How the cost of a mortgage has changed in the UK – from the 60s until now (2025)

https://hotminute.co.uk/2025/06/22/heres-how-the-cost-of-a-mortgage-has-changed-in-the-uk-from-th...
2•robtherobber•18m ago•0 comments

Claude, Gemini, and Copilot Got Hijacked

https://agentshield.pro/blog
1•eigenart•19m ago•0 comments

What's the Point of Hardbacks?

https://tomrowley.substack.com/p/whats-the-point-of-hardbacks
2•casca•19m ago•0 comments

The New Economics of WPF Migration: Why AI Agents Changed the Math

https://platform.uno/blog/the-new-economics-of-wpf-migration-why-ai-agents-changed-the-math/
1•mx2ei1•21m ago•0 comments

Slop Cop

https://awnist.com/slop-cop
3•ericHosick•22m ago•0 comments

ERC explains stricter application measures amid rising demand for funding

https://erc.europa.eu/news-events/news/erc-president-explains-stricter-application-measures-amid-...
1•ruieduardolopes•23m ago•1 comments

Plant a Thousand Flowers, Pull Ten Thousand Weeds

https://stripmall.software/blog/plant-a-thousand-flowers/
1•rmoskal•23m ago•0 comments

The Claude Coding Vibes Are Getting Worse

https://blog.matthewbrunelle.com/the-claude-coding-vibes-are-getting-worse/
2•birdculture•23m ago•0 comments

Rethinking AI TCO: Why Cost per Token Is the Only Metric That Matters

https://blogs.nvidia.com/blog/lowest-token-cost-ai-factories/
1•gmays•24m ago•0 comments

Agents that remember: introducing Cloudflare Agent Memory

https://blog.cloudflare.com/introducing-agent-memory/
2•tysont•24m ago•0 comments