Hacker News: reidbarber

New comment by reidbarber in "Build and Host AI-Powered Apps with Claude – No Deployment Needed"

reidbarber — Wed, 25 Jun 2025 18:23:42 +0000

The big feature here is that the shared artifacts can use the Claude API themselves (where usage is tied to the logged-in users of your shared artifact).

Reverse Engineering Claude Code

reidbarber — Thu, 08 May 2025 17:39:31 +0000

Article URL: https://www.reidbarber.com/blog/reverse-engineering-claude-code

Comments URL: https://news.ycombinator.com/item?id=43928894

Points: 1

# Comments: 0

New comment by reidbarber in "Ask HN: How much better are AI IDEs vs. copy pasting into chat apps?"

reidbarber — Thu, 08 May 2025 15:17:24 +0000

I've found that to be best for coding, especially when you're describing the exact behavior you want implemented. Plus the format it returns is very consistent (not the case in the various chat UIs).

New comment by reidbarber in "Ask HN: How much better are AI IDEs vs. copy pasting into chat apps?"

reidbarber — Thu, 08 May 2025 14:30:30 +0000

My current favorite (and free) workflow:

1. Drop project files into https://files2prompt.com to copy a prompt with all file contents

2. Paste into https://aistudio.google.com/ and set a low (or 0) temperature and low top_p

Since Gemini 2.5 Pro is free in AI Studio at the moment, and there's a 1M token limit, this works for most things I need. Cursor is better in some cases where I need a bunch of small edits across a bunch of different files.

Show HN: WebMarker – Mark web pages for use with vision-language models

reidbarber — Fri, 24 Jan 2025 17:59:39 +0000

WebMarker is a JavaScript library used for adding visual markers and labels to elements on a web page. This can be used for Set-of-Mark prompting, which improves visual grounding abilities of vision-language models such as GPT-4o, Claude 3.5, and Google Gemini 1.5. This library aims to:

- Improve LLM performance on vision tasks referencing web pages - Enable reliable web page interactions based on LLM responses

Comments URL: https://news.ycombinator.com/item?id=42815619

Points: 2

# Comments: 1

New comment by reidbarber in "Show HN: Repogather – copy relevant files to clipboard for LLM coding workflows"

reidbarber — Thu, 12 Sep 2024 22:32:45 +0000

Nice! I built something similar, but in the browser with drag-and-drop at https://files2prompt.com

It doesn’t have all the fancy LLM integration though.

Show HN: WebMarker – Mark web pages for use with large language models

reidbarber — Tue, 03 Sep 2024 13:58:33 +0000

WebMarker is a JavaScript library used for adding visual markers and labels to elements on a web page. This can be used for Set-of-Mark prompting, which improves the visual grounding abilities of vision-enabled large language models such as GPT-4o, Claude 3.5, and Google Gemini 1.5. This library aims to:

- Improve LLM performance on vision tasks referencing web pages - Enable reliable web page interactions based on LLM responses

Comments URL: https://news.ycombinator.com/item?id=41435017

Points: 1

# Comments: 0

Reverse Engineering Claude Artifacts

reidbarber — Mon, 24 Jun 2024 22:05:55 +0000

Article URL: https://www.reidbarber.com/blog/reverse-engineering-claude-artifacts

Comments URL: https://news.ycombinator.com/item?id=40781477

Points: 1

# Comments: 0

New comment by reidbarber in "Show HN: Tarsier – Vision utilities for web interaction agents"

reidbarber — Wed, 15 May 2024 22:47:33 +0000

Neat! Been building something similar to the tagging feature in Typescript: https://github.com/reidbarber/webmarker

The Python API on this is really nice though.

New comment by reidbarber in "Show HN: FileKitty – Combine and label text files for LLM prompt contexts"

reidbarber — Sun, 05 May 2024 21:59:25 +0000

Thanks, I appreciate it! I hope to keep adding features. If there are missing features that you'd use, feel free to leave them in a reply.

New comment by reidbarber in "Show HN: FileKitty – Combine and label text files for LLM prompt contexts"

reidbarber — Wed, 01 May 2024 22:14:16 +0000

Nice! I made something similar but for the browser recently: https://files2prompt.com

I think there some CLI tools out there as well.