Urgent.News

What's breaking now, across thousands of outlets.

Tech

The Design Logic Behind Eight Skills

In the previous post I said that the defenses we bought with real incidents eventually turned into eight concrete SKILLs. This post explains how to use those eight. Before I unpack them, one premise has to be settled: the tools serve the project specification document (ProjectSpec) , because that document is the first-class citizen of the whole system. It is not only a document asset for people…

The eight SKILLs are designed to support the project specification document, which serves as a machine contract for AI. Each SKILL covers a specific stage in the development process, from design to development to review. The first group of four SKILLs, known as the issue chain, focuses on maintaining the connection between conversation and the record.

The agenda established at the beginning of a discussion serves as the anchor, ensuring that discussions are properly concluded and tracked in the project document. These four SKILLs are typically used when a discussion spans multiple AI sessions or when a discussion needs to be archived after an abnormal interruption. The second group of four SKILLs, referred to as the delivery chain, aims to prevent design and implementation from becoming separate entities.

The spec-dev-loop SKILL ensures that code skeletons are frozen before implementation begins, while the spec-code-review SKILL addresses changes made during development that may cause the documentation to become outdated. The final SKILL, spec-code-smell, focuses on maintaining code quality by identifying potential issues early in the development process.

These eight SKILLs are not designed to replace human judgment or guarantee the correctness of the design. Instead, they provide a framework for ensuring that decisions made during the development process are properly documented and can be checked by the AI.

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

Read the original at dev.to →

More in Tech

Self-Hosting Supabase on VPS with Coolify v4

💡 Originally published on Bijak Technology Enterprise Journal with full interactive parameter benchmarks. Commercial Supabase Cloud Pro bills spike aggressively the moment your application crosses…

🌳 Pine Script into readable Python for Free

Download the converted 449 strategies and 471 indicators via this link I built a free compiler and translated over 1000 popular trading strategies and indicators from TradingView into Python source…

  • Free compiler translates 1,000+ TradingView strategies to Python
  • 384 strategies and 458 indicators achieve bit-for-bit consistency
  • Supports multiple exchanges via CCXT library for candlestick charting

Long context vs RAG: why search when a model reads it all?

For long context vs RAG, start with full-document input when one document fits and you need only a few answers; test retrieval for repeated questions over a large collection.

  • Long context allows processing entire document if within model's 128k-1M token window
  • RAG offers efficiency for large document collections by retrieving key passages
  • RAG risk lies in retrieval accuracy, potentially leading to incorrect answers

ag-ui: A Chinese Translation of the Agent-User Interaction Protocol

ag-ui-protocol/ag-ui has 20219 stars and is written in TypeScript. The Chinese translation at yangshun2005/ag-ui-cn was published on 2026-11-18 and currently has no stars of its own.

  • AG-UI protocol enables AI agents to interact with user-facing apps without request/response model
  • Chinese translation hosted at yangshun2005/ag-ui-cn, published 2026-11-18
  • Guide includes CLI scaffold and protocol example setup paths

More from Wednesday 18 November →