📰 Dev.to · sisyphusse1-ops

Articles from Dev.to · sisyphusse1-ops · 4 articles · Updated every 3 hours · View all reads

All ⚡ AI Lessons (26770) ArXiv cs.AI Dev.to AI Medium · AI Medium · Programming Medium · Machine Learning Medium · Cybersecurity

I shipped cc-audit as a GitHub Action. Now your CLAUDE.md gets linted on every PR.

Dev.to · sisyphusse1-ops 2d ago

I shipped cc-audit as a GitHub Action. Now your CLAUDE.md gets linted on every PR.

Three-line GitHub Action that lints your CLAUDE.md on every PR. Plus a companion action that installs the 12-rule baseline.

I scored 492 public CLAUDE.md files against a 12-rule baseline. Median: 3/12.

Dev.to · sisyphusse1-ops 2d ago

I scored 492 public CLAUDE.md files against a 12-rule baseline. Median: 3/12.

Scanned 492 real CLAUDE.md files. Median compliance with the 12-rule pack: 3/12. Here's what everyone skips and the 90-second fix.

I built a coding agent that runs on Gemma 4 — here's what 2B parameters can actually do

Dev.to · sisyphusse1-ops 2d ago

I built a coding agent that runs on Gemma 4 — here's what 2B parameters can actually do

A single-file Python CLI that turns Gemma 4 E2B (2B-effective) into a working coding agent. Rulebook-first, XML tool protocol, runs on a Raspberry Pi 5.

I read 31 pages of Anthropic prompting guidance so you don't have to — here's what actually changes with Claude 4.7

Dev.to · sisyphusse1-ops 🧠 Large Language Models ⚡ AI Lesson 2d ago

I read 31 pages of Anthropic prompting guidance so you don't have to — here's what actually changes with Claude 4.7

Claude Opus 4.7 follows prompts literally. Six shifts and a rewrite checklist to bring your 4.6-era prompts up to date.