Hacker News: dbreunig

Manage Your Agent's Loadout with Dr Skill

dbreunig — Mon, 27 Jul 2026 23:32:05 +0000

Article URL: https://www.dbreunig.com/2026/07/24/manage-your-agent-s-loadout-with-dr-skill.html

Comments URL: https://news.ycombinator.com/item?id=49077030

Points: 2

# Comments: 0

Show HN: Brew doctor` for Skills and MCP loadouts

dbreunig — Wed, 22 Jul 2026 15:50:45 +0000

Recently, I've had agents reach for the wrong skill or tool. This tends to happen in the middle of longer reasoning chains, which frustratingly means lots of chatter polluting the context before a correction can be issued.

To chase these issues down, I built `drskill`: a CLI tool that checks for conflicts and security issues among your agent's Skill and MCP loadout.

In addition to spotting conflicts that might trip up your agents, security alerts were added as well. Did an MCP provider silently add a tool due to a supply chain attack? Running `drskill` on bootup or during CI will flag those changes.

`drskill` works without access to an LLM, though connecting it to one will yield better conflict reports and allows for suggesting alternate, non-clashing descriptors.

A ledger is maintained globally and locally, tracking acknowledgements and preferences.

This is the type of tool that heavily benefits from contributions (your loadout is very different than mine!), so please file issues or submit PRs if you're interested.

Comments URL: https://news.ycombinator.com/item?id=49008729

Points: 2

# Comments: 0

New comment by dbreunig in "Show HN: I left Figma to build a diffusion-based UI design tool"

dbreunig — Tue, 21 Jul 2026 17:57:31 +0000

I got an early tip on this and used it to build a brand mood board and it crushed previous attempts with Claude. Highly recommend.

Understanding the Dynamics of the AI Ecosystem with Pace Layers

dbreunig — Sat, 04 Jul 2026 16:07:50 +0000

Article URL: https://www.dbreunig.com/2026/07/03/ai-ecosytem-pace-layers.html

Comments URL: https://news.ycombinator.com/item?id=48786426

Points: 3

# Comments: 1

The Problem is Prompt Debt: You can't be model agnostic and hand-tune prompts

dbreunig — Tue, 23 Jun 2026 16:20:28 +0000

Article URL: https://www.dbreunig.com/2026/06/22/the-problem-is-prompt-debt.html

Comments URL: https://news.ycombinator.com/item?id=48647349

Points: 4

# Comments: 0

When an agent can explain anything, what is the role of human-centric docs?

dbreunig — Sun, 31 May 2026 21:24:09 +0000

Article URL: https://www.dbreunig.com/2026/05/31/what-do-humans-need-from-docs.html

Comments URL: https://news.ycombinator.com/item?id=48349900

Points: 4

# Comments: 0

New comment by dbreunig in "Sam Altman's Business Dealings Under GOP Scrutiny Ahead of OpenAI's IPO"

dbreunig — Thu, 14 May 2026 15:51:16 +0000

Elizabeth Lopatto at The Verge makes a strong case we _do_ have proof that Musk is actively gathering and throwing fuel on the fire: https://www.theverge.com/ai-artificial-intelligence/929129/s...

> But the thing is, Molo doesn’t actually have to be good at this job, because the point of this trial isn’t to win — though I’m sure Musk wouldn’t mind a win. The point is to punish Altman, Brockman, and OpenAI. Musk has done that pretty thoroughly — reinforcing in the public’s mind that Altman is a liar and a snake. This morning, I read an exclusive in The Wall Street Journal that assorted Republican AGs and the House Oversight committee wanted to look into Sam Altman’s investments. References to the trial are peppered throughout the article.

Overfitting to First Party Harnesses

dbreunig — Sun, 10 May 2026 16:39:42 +0000

Article URL: https://www.dbreunig.com/2026/05/10/overfitting-the-harness.html

Comments URL: https://news.ycombinator.com/item?id=48085419

Points: 1

# Comments: 0

New comment by dbreunig in "Accelerating Gemma 4: faster inference with multi-token prediction drafters"

dbreunig — Tue, 05 May 2026 22:19:52 +0000

Among benchmarkers its a frequent topic. Qwen BURNS reasoning to get its scores.

Cybersecurity looks like proof of work now

dbreunig — Tue, 14 Apr 2026 18:08:14 +0000

Article URL: https://www.dbreunig.com/2026/04/14/cybersecurity-is-proof-of-work-now.html

Comments URL: https://news.ycombinator.com/item?id=47769089

Points: 562

# Comments: 214

Claude Code builds a system prompt

dbreunig — Sun, 05 Apr 2026 14:47:21 +0000

Article URL: https://www.dbreunig.com/2026/04/04/how-claude-code-builds-a-system-prompt.html

Comments URL: https://news.ycombinator.com/item?id=47650000

Points: 4

# Comments: 0

The Cathedral, the Bazaar, and the Winchester Mystery House

dbreunig — Fri, 03 Apr 2026 17:01:52 +0000

Article URL: https://www.oreilly.com/radar/the-cathedral-the-bazaar-and-the-winchester-mystery-house/

Comments URL: https://news.ycombinator.com/item?id=47629178

Points: 2

# Comments: 0

The 2nd phase of OSS in the agentic era: From clones to reimaginings

dbreunig — Thu, 02 Apr 2026 16:17:48 +0000

Article URL: https://www.dbreunig.com/2026/04/01/the-2nd-phase-of-agentic-development.html

Comments URL: https://news.ycombinator.com/item?id=47616459

Points: 2

# Comments: 0

The Cathedral, the Bazaar, and the Winchester Mystery House

dbreunig — Wed, 01 Apr 2026 14:13:34 +0000

Article URL: https://www.dbreunig.com/2026/03/26/winchester-mystery-house.html

Comments URL: https://news.ycombinator.com/item?id=47601194

Points: 190

# Comments: 66

New comment by dbreunig in "If DSPy is so great, why isn't anyone using it?"

dbreunig — Mon, 23 Mar 2026 16:41:07 +0000

Model testing and swapping is one of the surprises people really appreciate DSPy for.

You're right: prompts are overfit to models. You can't just change the provider or target and know that you're giving it a fair shake. But if you have eval data and have been using a prompt optimizer with DSPy, you can try models with the one-line change followed by rerunning the prompt optimizer.

Dropbox just published a case study where they talk about this:

> At the same time, this experiment reinforced another benefit of the approach: iteration speed. Although gemma-3-12b was ultimately too weak for our highest-quality production judge paths, DSPy allowed us to reach that conclusion quickly and with measurable evidence. Instead of prolonged debate or manual trial and error, we could test the model directly against our evaluation framework and make a confident decision.

https://dropbox.tech/machine-learning/optimizing-dropbox-das...