Urgent.News

What's breaking now, across thousands of outlets.

AI

State of AI Instruction Quality: /init Files Score Higher

The CLAUDE.md files that started from Claude Code's /init score higher than the rest, at every file size. You can tell them by the line /init puts at the top: This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository. We scored 8,220 CLAUDE.md files from the 10,000 most-starred repos on GitHub for the Q3 report . One in five has that line. On ails check…

The /init files, which are CLAUDE.md files containing guidance for Claude Code, tend to score higher than other CLAUDE.md files, regardless of their size. The /init line at the top of these files indicates that they provide guidance to Claude Code on working with code in the repository. According to the Q3 report, 8,220 out of the 10,000 most-starred repositories on GitHub have a /init file, and a significant portion of these files score well on the Q3 evaluation scale.

These /init files have a higher median score of 8.4, compared to 7.7 for the other files. Six in ten of the /init files score 8 or above, while fewer than half of the other files reach the same threshold. The /init files' structure, which includes a map of the repository, command instructions in code blocks, and a short list of rules, also contributes to their higher scores. When the /init line is removed, the file scores lower, but maintaining the /init file's structure can still yield a high score.

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 AI

FreelanceFlow AI: Turn Messy Client Briefs into Proposals with a Local LLM

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built FreelanceFlow AI turns a messy client brief into a structured, editable project proposal, and it…

  • FreelanceFlow AI simplifies proposal creation for freelancers
  • Tool extracts key client brief info locally, ensuring data security
  • Built with Python, FastAPI, React, Vite, and Ollama for local inference

ScopeGuard: Helping Freelancers Handle Scope Creep with Open-Weight AI

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built Freelancers know the situation: a client asks for one more feature, another revision, or an…

  • ScopeGuard helps freelancers manage scope creep in client relationships.
  • Analyzes client requests against agreed-upon project scope.
  • Generates professional responses using open-weight AI model.

More from Wednesday 14 October →