Urgent.News

What's breaking now, across thousands of outlets.

AI

I got tired of typing "do 1 and 3, skip 2" to Claude Code, so I built an approve/reject board

You ask Claude Code for ten improvements to your project. It answers with a tidy numbered list. Then the real work starts: you scroll, you type "do 1 and 3, skip 2 because we tried that, maybe 5 later," and you hope the agent maps your sentence back to the right items. Next week, in a new session, it proposes number 2 again. So I built ar-board , a Claude Code plugin that turns the list into…

Claude Code users have found the process of selecting improvements from a list unproductive, as they have to manually type "do" before each number, even if they rejected previous suggestions. This repetitive task led to the creation of an approve/reject board, called ar-board, which turns the list into cards on a local page. Using this board, users can quickly approve, reject, or mark items for later consideration, with their decisions reflected in a preferences file for future interactions with Claude Code.

The plugin is simple, requiring only one Python file and standard library dependencies, and is licensed under MIT. It supports both English and Portuguese interfaces and runs securely on localhost, with additional security measures such as token-based authentication and restricted access from other domains. The plugin is designed to be unobtrusive, allowing users to customize card labels and enforce second approvals for risky changes.

While it is currently limited to single-user use on localhost, future updates may extend its functionality to remote environments. The ar-board tool is part of the Capybara Toolbox suite, an MIT-licensed collection of Claude Code plugins.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at dev.to →

More in AI

The Aircraft Manual Prompt That Cuts AI Waffle: ASD-STE100 for ChatGPT and Claude

You ask a chatbot a simple question and get four paragraphs back. The first one repeats your question. The second one hedges.

  • Chatbot responses often feel bloated and confusing.
  • Andrej Karpathy proposed using Simplified Technical English (STE) for clearer AI outputs.
  • STE rules force concise, active voice, and single meaning words to reduce variability.

Are We Looking for AI Control in the Wrong Place?

Billions of dollars. The sharpest minds in computing. Entire industries built around one question: How do we make an LLM do the right thing—predictably, repeatedly, every time? More training.

  • Industry focuses on refining probabilistic models rather than questioning authority.
  • Artificial Deterministic Intelligence (ADI) suggests removing model's decision-making power.
  • ADI shifts control back to humans, making decisions auditable and owned by people.

Building a Portable AI Collaboration Protocol With Explicit Limits

Building a Portable AI Collaboration Protocol With Explicit Limits The Artemis / Solomon’s Staff protocol v5.6.0 is a machine-readable JSON document for carrying a user-approved collaboration context…

  • Artemis/Solomon s Staff protocol version 5.6.0 facilitates user-approved AI collaboration
  • Protocol records communicable profile, project context with provenance
  • Protocol only activates when explicitly loaded by AI host, not controlling unrelated systems

Trojan Horse: AI agents that call mechanics for you, and how Kane CLI caught our bug

The problem Your car breaks down at night. You search for mobile mechanics and call them one by one: who’s open, who’s close, who won’t overcharge? That’s exactly when time matters most.

  • Trojan Horse AI app connects drivers with mechanics via mobile
  • Discovery, Caller, and Callback agents handle booking process
  • Kane CLI tests reveal bug where app incorrectly showed mechanic as available

More from Sunday 25 October →