frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Why XML Tags Are So Fundamental to Claude

https://glthr.com/XML-fundamental-to-Claude
31•glth•1h ago

Comments

TheJoeMan•1h ago
That first image, “Structure Prompts with XML”, just screams AI-written. The bullet lists don’t line up, the numbering starts at (2), random bolding. Why would anyone trust hallucinated documentation for prompting? At least with AI-generated software documentation, the context is the code itself, being regurgitated into bulleted english. But for instructions on using the LLM itself, it seems pretty lazy to not hand-type the preferred usage and human-learned tips.
Calavar•1h ago
It looks like a screenshot from the Claude desktop app, so I don't think the author is trying to disguise the AI origin of the marerial
rafram•54m ago
No, it’s two screenshots from Anthropic documentation, stitched together: https://platform.claude.com/docs/en/build-with-claude/prompt...

The post even links to that page, although there’s a typo in the link.

dmd•16m ago
They're not even stitched together ; there's just no padding between the two images.
glth•11m ago
Author here: I have just fixed the typo. Thank you.

And yes, these are screenshots from Anthropic’s documentation.

wolttam•1h ago
Anthropic’s tool calling was exposed as XML tags at the beginning, before they introduced the JSON API. I expect they’re still templating those tool calls into XML before passing to the model’s context
pocketarc•41m ago
Yeah like I remember prior to reasoning models, their guidance was to use <think> tags to give models space for reasoning prior to an answer (incidentally, also the reason I didn't quite understand the fuss with reasoning models at first). It's always been XML with Anthropic.
wolttam•10m ago
Exactly the same story here. I still use a tool that just asks them to use <think> instead of enabling native reasoning support, which has worked well back to Sonnet 3.0 (their first model with 'native' reasoning support was Sonnet 3.7)
imglorp•34m ago
A very minor porcelain on some of the agent input UX could present this structure for you. Instead of a single chat window, have four: task, context, constraints, output format.

And while we're at it, instead of wall-of-text, I also feel like outputs could be structured at least into thinking and content, maybe other sections.

Zebfross•33m ago
I thought the goal was minimal instruction to let Claude determine the best way to solve the problem. Not adding this to my workflow anytime soon.
esafak•27m ago
This sounds like something for harnesses, not end users. Are they really expecting us to format prompts as XML??

AArch64 Bitfield Move (BFM) Instruction

https://nemanjatrifunovic.substack.com/p/aarch64-bitfield-move-bfm-instruction
1•ingve•2m ago•0 comments

World-first gigabit laser link between aircraft and geostationary satellite

https://www.esa.int/Applications/Connectivity_and_Secure_Communications/World-first_gigabit-per-s...
1•giuliomagnifico•2m ago•0 comments

Show HN: I made a iron dome game

https://deployclaw.com/games/iron-dome
2•manishipsfast•3m ago•0 comments

Making a Type Documentary

https://jonathanhoefler.com/articles/making-a-type-documentary
1•simonebrunozzi•5m ago•0 comments

Evil in the West Bank – David Shulman

https://www.nybooks.com/articles/2026/03/12/evil-in-the-west-bank-david-shulman/
3•abdelhousni•8m ago•0 comments

Show HN: Polars.net – 36x faster than pyspark on deltalake

https://github.com/ErrorLSC/Polars.NET
1•errorlsc•10m ago•1 comments

Agent Relay – open-source rails to have agents communicate

https://twitter.com/willwashburn/status/2027536741456863504
1•predogger•10m ago•0 comments

We've got 10 sales from Hacker News

https://www.founderspace.work
1•VladCovaci•11m ago•1 comments

McDonald's is not lovin' your bigmac, happymeal, and mcnuggets passwords

https://www.theregister.com/2026/02/02/mcdonalds_password_advice/
1•gnabgib•11m ago•0 comments

10k+ Funny Quotes

https://gagzap.com
1•hackerbeat•13m ago•1 comments

Show HN: OpenTT – a self-hosted tool for managing table tennis leagues

https://github.com/tradicije/ttopen
1•tradicije•13m ago•0 comments

Rust Fuzz Book

https://rust-fuzz.github.io/book/
1•vinhnx•15m ago•0 comments

Anthropic's Killer-Robot Dispute with The Pentagon

https://www.theatlantic.com/technology/2026/03/inside-anthropics-killer-robot-dispute-with-the-pe...
2•spenvo•15m ago•1 comments

LLMs not = Security Products

http://hooked-on-mnemonics.blogspot.com/2026/02/llms-security-products.html
1•__alexander•15m ago•1 comments

Iran's Missile Barrage Tests Whether US Has Enough Interceptors

https://www.bloomberg.com/news/articles/2026-02-28/iranian-missile-attacks-set-to-strain-us-inter...
2•ParentiSoundSys•17m ago•0 comments

An Interview from 2036 with Elon Musk, Jeff Bezos and Sam Altman

https://www.aicandy.be/giorgio-1
2•wodniok•21m ago•0 comments

Ruby -run, utilities to replace common Unix commands

https://github.com/ruby/un
2•Alifatisk•25m ago•0 comments

Project Zero Trust: A story about a strategy for aligning security and business

https://cybercanon.org/project-zero-trust-a-story-about-a-strategy-for-aligning-security-and-the-...
3•teleforce•27m ago•0 comments

The Hard Truth About AI Agents: What We Learned Building in Open Source

https://alexchao.substack.com/p/the-hard-truth-about-ai-agents-what
1•alexchaomander•28m ago•1 comments

AI utility is perceived uniquely by programmers, scientists, and writers

1•kokhanserhii•30m ago•0 comments

Chief Mouser to the Cabinet Office

https://en.wikipedia.org/wiki/Chief_Mouser_to_the_Cabinet_Office
1•alabhyajindal•30m ago•0 comments

Show HN: Interpreted Art – Explore Symbolism in the Arnolfini Portrait

https://interpreted.art/art/MlLrLHxRKZHdQsGkv3RT
1•sailorganymede•31m ago•0 comments

Did a Restorer Paint Italian Prime Minister Giorgia Meloni into a Church Fresco?

https://news.artnet.com/art-world/church-fresco-giorgia-meloni-2742134
1•gnabgib•34m ago•0 comments

Show HN: I stopped wireframing pages – I design SaaS products as visual sitemaps

3•epic_ai•34m ago•5 comments

How I generate podcast-style audio overviews [video]

https://www.youtube.com/watch?v=mLiXIQjQ5CU
1•digitalhobbit•35m ago•1 comments

Show HN: Hmem v2 – Persistent hierarchical memory for AI agents (MCP)

https://github.com/Bumblebiber/hmem
1•Bumblebiber•35m ago•1 comments

AI contributed "zero" to U.S. economic growth, Goldman Sachs has calculated

https://www.washingtonpost.com/technology/2026/02/23/ai-economic-growth-gdp-mirage/
2•coliveira•40m ago•0 comments

Show HN: Glass box governance for multi-agent AI coding workflows

https://github.com/Vinix24/vnx-orchestration
1•vincentvandeth•42m ago•1 comments

Show HN: Plagiarismremover.ai – A Free AI Plagiarism Remover

https://plagiarismremover.ai/
2•PlagiarismRem•45m ago•0 comments

What I learned building a Multi-Agent System

https://davide.im/posts/multi-agent-system/
2•vector3•46m ago•0 comments