AI Coding Config
Claude Code plugin marketplace with 18 commands, 24 agents, and 33 coding rules
Install
/plugin marketplace add https://github.com/TechNickAI/ai-coding-configInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →AI Coding Configuration
Curated commands, agents, and rules for Claude Code, Cursor, Windsurf, and Cline.
What This Is
A shared configuration that works across AI coding tools. Commands automate workflows (PR handling, debugging, session management). Agents specialize in specific review types (security, performance, UX). Rules encode your coding standards so AI follows your patterns.
Quick Start
One-line install
Run this in your terminal — it auto-detects your tools (Claude Code, Cursor, etc.) and sets everything up:
curl -fsSL https://raw.githubusercontent.com/TechNickAI/ai-coding-config/main/scripts/bootstrap.sh | bash
Then configure for your project:
/ai-coding-config
Manual install (Claude Code)
If you prefer to install manually, run these in Claude Code:
/plugin marketplace add https://github.com/TechNickAI/ai-coding-config
/plugin install ai-coding-config
Todo Persistence Across Compaction
The problem: Claude Code's context compaction summarizes conversation history to stay within token limits. When this happens, your todo list vanishes - you lose track of what you were working on.
The solution: This plugin automatically saves todos to disk via hooks. After compaction, restore them:
cat ~/.claude/projects/$(echo $PWD | sed 's|/|-|g')/todos.md
Then use TodoWrite to restore your progress. Your todos survive compaction, session
restarts, and even --resume across days.
Example Usage
# Triage PR bot comments (fixes real issues, declines nitpicks)
/address-pr-comments
# Load coding standards for current task
/load-rules
# Autonomous task execution: describe what you want, get a PR
/autotask "add user settings page with dark mode toggle"
What's Included
| Type | Count | Purpose |
|---|---|---|
| Commands | 18 | Automate workflows |
| Agents | 24 | Specialized assistants |
| Skills | 6 | Autonomous capabilities |
| Rules | 33 | Coding standards |
| Personalities | 7 | Communication styles |
Commands
Slash commands that automate real workflows. Type /command-name to invoke.
Autonomous Development
/autotask "description" - Describe what you want, get a PR. Creates branch,
implements, writes tests, handles bot feedback.
/autotask "add user settings page with dark mode toggle"
/troubleshoot - Connects to Sentry/HoneyBadger, analyzes errors, fixes bugs in
parallel worktrees, submits PRs with root cause analysis.
/verify-fix - Confirms fixes actually work before claiming success. Runs tests,
checks live behavior.
PR & Code Review
/address-pr-comments - Triages bot comments: fixes real issues, declines nitpicks,
iterates until merge-ready.
/multi-review - Runs multiple specialized reviewers in parallel. Security, logic,
performance, style - all at once.
Context & Session Management
/session save|resume|list - Save your context, decisions, and progress. Resume
exactly where you left off - even in a new conversation.
/session save "auth-refactor" # Save current session
/session resume # Resume where you left off
Similar plugs
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
Documented system prompts from Anthropic - Claude Fable 5.1, Opus 5.5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro,…
Wrap Antigravity, ChatGPT Codex, Claude Code, Grok Build, Muse Code, Devin as an OpenAI/Gemini/Claude/Codex compatible API service, allowing you to enjoy the free Gemini…