Hacker News

I bought a WCH BLE Analyzer PRO, only to realize it doesn't have Linux support. Using the windows binaries, i reverse engineered their USB implementation, and vibe-coded a capture tool together in less than 24 hours.

Comments URL: https://news.ycombinator.com/item?id=47091034

Points: 1

# Comments: 1

Article URL: https://github.com/exitxio/claude-code-web

Comments URL: https://news.ycombinator.com/item?id=47091003

Points: 1

# Comments: 1

Article URL: https://www.zdnet.com/article/red-hat-digital-sovereignty-toolkit/

Comments URL: https://news.ycombinator.com/item?id=47090983

Points: 1

# Comments: 0

Article URL: https://connectsci.au/wr/article/53/2/WR25088/268264/Absence-or-avoidance-White-shark-response-to

Comments URL: https://news.ycombinator.com/item?id=47090971

Points: 1

# Comments: 0

Article URL: https://en.wikipedia.org/wiki/Information_Processing_Language

Comments URL: https://news.ycombinator.com/item?id=47090970

Points: 1

# Comments: 0

Article URL: https://www.terrygodier.com/current

Comments URL: https://news.ycombinator.com/item?id=47090963

Points: 1

# Comments: 0

Article URL: https://xcancel.com/kimmonismus/status/2024502735584780593

Comments URL: https://news.ycombinator.com/item?id=47090959

Points: 3

# Comments: 2

Article URL: https://www.youtube.com/watch?v=nufOQWBmwpk

Comments URL: https://news.ycombinator.com/item?id=47090947

Points: 2

# Comments: 1

Article URL: https://ccbench.org/

Comments URL: https://news.ycombinator.com/item?id=47090946

Points: 1

# Comments: 0

Article URL: https://lexifina.com/blog/memory-for-legal-ai

Comments URL: https://news.ycombinator.com/item?id=47090931

Points: 1

# Comments: 0

Article URL: https://arxiv.org/abs/2602.16800

Comments URL: https://news.ycombinator.com/item?id=47090929

Points: 1

# Comments: 0

Article URL: https://github.com/trezm/superposition

Comments URL: https://news.ycombinator.com/item?id=47090916

Points: 1

# Comments: 1

Live: https://subconscious-scheduler.vercel.app/dashboard/tasks Repo: https://github.com/ostepan8/subconscious-scheduler

I built this because I wanted a daily briefing on my sports teams and stocks without checking five different apps. You pick your topics, set a schedule, and an agent researches and writes the email for you.

Stack: Convex for backend/scheduling, Resend for email delivery, Claude for agent logic. Setup takes ~30 seconds through the UI, or describe what you want to the chat agent and it configures itself. Self-hostable, but bring your own API keys (live is completely free).

Comments URL: https://news.ycombinator.com/item?id=47090908

Points: 1

# Comments: 0

Article URL: https://quillette.com/2026/02/16/keep-calm-and-adapt-ai-matt-shumer-automation/

Comments URL: https://news.ycombinator.com/item?id=47090872

Points: 1

# Comments: 0

I built tuber to get information from YouTube up to 100x faster.

Two core features:

- If a video title is a question, tuber gives a direct answer first. - Then it generates a timestamped summary with Claude, so you can jump to the exact parts that matter.

Example:

- Video Title: "Did People Used To Look Older?" - Tuber comment at the bottom: "People in old photos genuinely appear older than their current-day equivalents — but this "retrospective aging" effect is both real and illusory."

Enjoy tuber!

Comments URL: https://news.ycombinator.com/item?id=47090870

Points: 1

# Comments: 0

Article URL: https://www.hyperagent.com/

Comments URL: https://news.ycombinator.com/item?id=47090865

Points: 1

# Comments: 0

Article URL: https://vitalize.care/blog/from-supabase-to-planetscale

Comments URL: https://news.ycombinator.com/item?id=47090202

Points: 1

# Comments: 0

Prothon is a Python project generator that scaffolds a uv-based project with eight quality tools (ruff, ty, pytest, hypothesis, mutmut, bandit, vulture, complexipy) and a documentation-driven workflow for AI coding agents.

The problem it solves: AI assistants lose context between sessions and drift from your decisions as context windows fill up. Prothon addresses this with three ideas:

1. A three-level doc hierarchy (SPEC, DESIGN, PATTERNS) where each level scopes a single concern and higher always overrides lower. The AI implements from these docs, not from memory. Each doc-writing step is a dedicated agent that refuses content belonging at another level.

2. Automated agents that keep things consistent: one cross-checks the three docs for contradictions, one generates reference skills from live documentation for your stack, one verifies code implements what the docs describe.

3. A "promise" system for execution. Before each task, the agent declares what files it will touch, success criteria, and expected line counts. The promise is verified against the actual git diff. This makes context drift measurable rather than something you notice after the fact.

CLI:

``` prothon new # scaffold from template prothon init # add to existing project prothon spec # requirements Q&A prothon design # architecture Q&A prothon patterns # conventions Q&A prothon execute # plan, promise, build prothon compliance # verify code vs docs ```

Built on copier, so existing projects can pull upstream template improvements. Currently targets Claude Code, designed to be agent-agnostic.

https://prothon.dev https://github.com/jackedney/prothon

Comments URL: https://news.ycombinator.com/item?id=47090181

Points: 1

# Comments: 0

Article URL: https://gomakethings.com/who-stole-your-cookie/

Comments URL: https://news.ycombinator.com/item?id=47090178

Points: 1

# Comments: 0

Article URL: https://taylor.town/base-units

Comments URL: https://news.ycombinator.com/item?id=47090177

Points: 1

# Comments: 0