OPERATIONAL·247 items indexed·Last sync: 7:00 AM PST·Next sync: Tomorrow 7:00 AM PST·2 sources activeOPERATIONAL·247 items indexed·Last sync: 7:00 AM PST·Next sync: Tomorrow 7:00 AM PST·2 sources active

Discover & master the latest AI coding tools

Daily curated skills, hooks, techniques, and workflows — primarily for Claude Code, plus Codex CLI and the broader AI ecosystem.

Daily Briefing

MONDAY, SEPTEMBER 14, 2026

Hooks, Multi-Agent Pipelines & the Art of Asking Better Questions

Today's intelligence covers three themes dominating the AI tooling conversation: Claude Code's hook system for automated quality gates, the emerging multi-agent orchestration pattern that separates planning from execution, and the surprisingly powerful impact of negative examples in prompt engineering. Plus: a deep dive into prompt caching economics that most teams are leaving on the table.

NEW TODAY

TOTAL INDEXED

247

SOURCES

prompt

Ultra-Compressed CLAUDE.md System Prompt for Maximum Context Efficiency

hook

Pre-Tool-Call Hooks: Auto-Lint Before Every File Write

workflow

Automated PR Review Workflow with GitHub Actions + Claude API

technique

Agentic Test-Driven Development: Write Tests First, Let Claude Implement

RECENTLY ADDED

VIEW ALL

prompt

Intermediate

Ultra-Compressed CLAUDE.md System Prompt for Maximum Context Efficiency

A structured CLAUDE.md template using XML-like shorthand notation that communicates your project architecture, conventions, and preferences to Claude Code in under 300 tokens — compared to the typical 800+. Uses abbreviation schemas and tiered priority markers.

Custom Slash Commands in Claude Code for Repetitive Workflows

Claude Code supports custom slash commands stored in .claude/commands/*.md. Each file becomes a reusable command — perfect for PR descriptions, commit messages, and code review workflows that you run daily.

Pre-Tool-Call Hooks: Auto-Lint Before Every File Write

Claude Code hooks let you intercept tool calls before and after execution. A PreToolUse hook on the Write tool can run ESLint/Prettier before Claude writes a file, ensuring every AI-generated file is already formatted — eliminating the "format the file" back-and-forth.

CodeGPT VS Code Plugin: In-Editor AI Chat with Codebase Context

CodeGPT is a VS Code extension that embeds AI chat directly in your editor sidebar with full codebase indexing. Unlike GitHub Copilot Chat, it supports multiple LLM backends (OpenAI, Anthropic, Ollama) and lets you create "agents" with custom system prompts per project.

Chain-of-Verification Prompting: Eliminate AI Hallucinations in Research Tasks

Chain-of-Verification (CoVe) is a prompting technique where you ask the AI to generate an answer, then independently generate verification questions, answer them without seeing the original response, and finally produce a revised answer. Studies show 30-40% reduction in factual errors.

Automated PR Review Workflow with GitHub Actions + Claude API

A GitHub Actions workflow that automatically reviews every pull request using the Claude API. It reads the diff, runs it through a structured review prompt, and posts inline comments on specific lines — catching logic errors, security issues, and style violations before human review.

Claude Code

GitHub