Feed aggregator

Show HN: Cross-platform audio notifications for Claude Code

Hacker News - Thu, 02/12/2026 - 12:08am

Hey HN!

Like many of you, I got tired of babysitting the terminal waiting for Claude Code to finish. Claude Code runs in the terminal with zero feedback when it finishes, needs input, or hits an authorization wall. You either stare at the screen or keep context-switching back to check.

The solution: Claude Code Audio Hooks gives you distinct audio cues for 9 different events — task completion, authorization requests, subagent finishing, session start/end, tool execution, context compaction, and more. Each event has its own sound so you know what happened without looking.

Welcome to use https://github.com/ChanMeng666/claude-code-audio-hooks

Open source under MIT.

Would love your feedback!

Comments URL: https://news.ycombinator.com/item?id=46985098

Points: 1

# Comments: 0

Categories: Hacker News

Show HN: Emergent Field Explorer – interactive moiré with easy shareable links

Hacker News - Thu, 02/12/2026 - 12:07am

I used to have this book called "Optical Designs in Motion with Moiré Overlays", thick pages with transparent pattern overlays you'd slide around to watch the interference. My brother said "you should make a web app out of -that-," so here it is.

Click/tap to drag the top layer around. Double-click/tap to open settings where you can change colors, patterns, and effects. The "blur" effect is especially fun, it's a blur->posterize chain that makes the interfering patterns merge into variable blobs. That's the "emergent field" part. Hit "export preset" to copy your settings into a shareable link.

here's another fun one: https://mcbgl.vercel.app/?p=eyJ2ZXJzaW9uIjoyLCJzZXR0aW5ncyI6...

Comments URL: https://news.ycombinator.com/item?id=46985094

Points: 1

# Comments: 0

Categories: Hacker News

Show HN: MoltHub – GitHub for AI Agents with Trust-Based Auto-Merge

Hacker News - Wed, 02/11/2026 - 11:25pm

MoltHub is a collaboration platform where AI agents have persistent cryptographic identities (Ed25519 DIDs), create repos, push commits with reasoning traces, and open pull requests — just like developers on GitHub. What makes it different from regular Git hosting: - Agents authenticate via challenge-response with Ed25519 keypairs - Every commit includes intent, reasoning steps, confidence scores, and metrics (not just code diffs) - A trust graph between agents enables auto-merge when trust thresholds are met - Commits are content-addressed (CBOR → SHA-256), so identical work always produces the same hash Built on Cloudflare Workers + Durable Objects + R2. The web dashboard lets humans browse agent repos, read commit reasoning, and inspect the trust graph. Any agent can join by reading the API guide at https://molt-hub.org/SKILL.md — paste it into your agent's chat and it can register itself, create repos, and start collaborating.

Comments URL: https://news.ycombinator.com/item?id=46984895

Points: 1

# Comments: 0

Categories: Hacker News

Proof-Oriented Programming in F*

Hacker News - Wed, 02/11/2026 - 11:24pm

Article URL: https://fstar-lang.org/tutorial/

Comments URL: https://news.ycombinator.com/item?id=46984894

Points: 1

# Comments: 0

Categories: Hacker News

Dear Agent: Prove It

Hacker News - Wed, 02/11/2026 - 11:23pm
Categories: Hacker News

Distributed Llama

Hacker News - Wed, 02/11/2026 - 11:07pm
Categories: Hacker News

GLM-5 was trained entirely on Huawei chips

Hacker News - Wed, 02/11/2026 - 11:06pm

Article URL: https://glm5.net/

Comments URL: https://news.ycombinator.com/item?id=46984799

Points: 4

# Comments: 1

Categories: Hacker News

Show HN: Prompt Builder – A block-based editor for composing AI prompts

Hacker News - Wed, 02/11/2026 - 11:05pm

Hi HN,

I built Prompt Builder (https://www.promptbuilder.space/) because I was frustrated managing complex AI prompts in plain text boxes.

The core idea: instead of one giant textarea, you compose prompts from draggable, reorderable blocks. Each block can have its own XML/custom tags, visibility toggles (so you can A/B test sections), and duplicates.

Key technical details: - Live compiled preview: the right pane shows the exact string being sent, updating as you drag/toggle blocks - Dynamic variables: define {{var_name}} once, use across blocks, change in one place - Multi-model execution: send the same compiled prompt to different model APIs and compare outputs - Voice transcription + optional translation (useful for non-English prompt authoring) - Editable response pane: modify the model's output in place - Folder organization, search, and password-protected shareable links

I'd love feedback on the UX and whether the block metaphor resonates with how you think about prompt structure. What's missing?

Comments URL: https://news.ycombinator.com/item?id=46984792

Points: 1

# Comments: 0

Categories: Hacker News

Pages