My Curated AI Programming Toolchain: Docs, Projects, and Tools Index

Hello, I'm James. After using AI programming tools for a while, my bookmarks filled with official docs, open‑source projects, research papers, and community tools became hard to locate. I reorganized the most frequently used resources by topic and share them as a backup and a reference for anyone working with AI programming.

01 | CodeBuddy Official Documentation

Plan Mode – https://www.codebuddy.ai/docs/zh/ide/Features/Plan-Mode – IDE Plan feature documentation

CODEBUDDY.md – https://www.codebuddy.ai/docs/zh/cli/best-practices#%E7%BC%96%E5%86%99%E6%9C%89%E6%95%88%E7%9A%84-codebuddy-md – Guide for writing effective CODEBUDDY.md files

Hooks – https://www.codebuddy.ai/docs/zh/cli/hooks – Hook configuration documentation

MCP – https://www.codebuddy.ai/docs/zh/cli/mcp – Model Context Protocol integration

Commands – https://www.codebuddy.ai/docs/zh/cli/commands – Custom command documentation

Subagents – https://www.codebuddy.ai/docs/zh/ide/Features/Subagents – Sub‑agent functionality

Sub‑agent Advanced Usage – https://www.codebuddy.ai/docs/zh/cli/sub-agents#%E9%AB%98%E7%BA%A7%E7%94%A8%E6%B3%95 – Details on Sub‑agent and Background Agent

Remote Control – https://www.codebuddy.ai/docs/zh/cli/wecom-bot-setup – Enterprise WeChat bot remote control

Memory System – https://www.codebuddy.ai/docs/zh/cli/memory – Cross‑session memory (Auto Memory)

Skills – https://www.codebuddy.cn/docs/ide/Features/Skills – Skills capability packs

Agent Teams – https://www.codebuddy.ai/docs/zh/cli/agent-teams#%E4%BD%95%E6%97%B6%E4%BD%BF%E7%94%A8-agent-teams – Multi‑agent collaboration

02 | Tools & Projects

Happy CLI – https://github.com/slopus/happy – Mobile AI programming tool

Paseo – https://paseo.sh/ – Seamless PC‑to‑mobile agent management

Ralph – https://github.com/snarktank/ralph – Self‑contained AI loop agent

Ralph‑wiggum – https://github.com/anthropics/claude-code/blob/main/plugins/ralph-wiggum/README.md – Claude Code plugin version of Ralph

Ralph‑orchestrator – https://github.com/mikeyobrien/ralph-orchestrator – Orchestrator for Ralph agents

Vibe‑Kanban – https://github.com/BloopAI/vibe-kanban – Visual task board

Agent Template Marketplace – https://app.aitmpl.com/agents – Sub‑agent scenario templates

OpenSpec – https://github.com/Fission-AI/OpenSpec/blob/main/docs/workflows.md – Specification‑driven development framework (workflow docs)

OpenSpec Customization – https://github.com/Fission-AI/OpenSpec/blob/main/docs/customization.md – Custom configuration for OpenSpec

Compound Engineering – https://github.com/EveryInc/compound-engineering-plugin – Plan→Work→Review→Compound workflow

Claude‑Mem – https://github.com/thedotmack/claude-mem – Claude cross‑session memory

Agent of Empires – https://github.com/njbrake/agent-of-empires – TUI/Web multi‑agent session manager supporting Claude Code, Codex, Gemini CLI, etc.

flow‑kit – https://github.com/rihebty/flow-kit – Markdown‑only AI programming best‑practice flow (no runtime dependencies)

03 | Agent Monitoring & Notifications

Vibe Island – https://vibeisland.app/ – Desktop task‑completion notifications (paid)

Open Vibe Island – https://github.com/Octane0411/open-vibe-island – Open‑source, macOS‑native alternative supporting multiple agents

claude‑notifications‑go – https://github.com/777genius/claude-notifications-go – Automatic long‑task completion alerts (Go implementation)

iTerm2 System Notifications – https://code.claude.com/docs/en/terminal-config#iterm-2-system-notifications – Claude Code terminal notification configuration

Selection Advice: For personal use, the free Open Vibe Island is sufficient; for team or multi‑platform scenarios, the paid Vibe Island offers a more complete experience.

04 | Code Review Related

code‑review.md – https://github.com/anthropics/claude-plugins-official/blob/main/plugins/code-review/commands/code-review.md – Official Code Review plugin

code‑simplifier agent – https://github.com/anthropics/claude-plugins-official/blob/main/plugins/code-simplifier/agents/code-simplifier.md – Code simplification agent

Context7 – https://context7.com/ – Latest documentation context assistance for reviews

05 | Skills Ecosystem

skills.sh – https://skills.sh/vercel-labs/skills/find-skills – Skills discovery and installation platform

skillsmp.com – https://skillsmp.com/ – Skills community marketplace

www.skills.sh – https://www.skills.sh/ – Official Skills website entry point

06 | Workflow & Git

Running multiple Agent instances in parallel relies heavily on Git Worktree.

Git Worktree – https://git-scm.com/docs/git-worktree – Official documentation for multiple worktrees

Git Worktrees Parallel – https://code.claude.com/docs/en/common-workflows#run-parallel-claude-code-sessions-with-git-worktrees – Parallel Claude Code sessions with Git worktrees

Cursor Worktrees – https://cursor.com/docs/configuration/worktrees – Cursor multi‑branch configuration

OpenAI Codex Worktrees – https://developers.openai.com/codex/app/worktrees/ – Codex app best‑of‑N worktree strategy

07 | AI Builder Information Flow

follow‑builders – https://github.com/zarazhangrui/follow-builders – AI builders digest: tracks top builder dynamics + podcasts, filters noise, supports bilingual pushes to Telegram/Discord

Karpathy Gist – https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f – Andrej Karpathy’s personal notes on LLM/Vibe coding workflows

08 | Community Knowledge Base & Standards

Awesome Claude Code – https://github.com/hesreallyhim/awesome-claude-code – Best practices, CLAUDE.md templates, toolchain collection

Awesome MCP Servers – https://github.com/punkpeye/awesome-mcp-servers – Indexed MCP server community

Prompt Engineering Guide – https://www.promptingguide.ai/zh – Systematic prompt‑engineering guide (Chinese version)

Google JavaScript Style Guide – https://google.github.io/styleguide/jsguide.html – Industry‑standard JS style guide

Airbnb JavaScript Style Guide – https://github.com/airbnb/javascript – Widely referenced JS coding conventions

MCP Official Docs – https://modelcontextprotocol.io/docs – Model Context Protocol official documentation

09 | Research & Papers

Politeness vs Accuracy in Prompts – https://arxiv.org/abs/2510.04950 – Study showing surprising impact of rude vs polite prompts on LLM accuracy

Installation Command Reminder

# OpenSpec CLI
npm install -g @fission-ai/openspec@latest

Summary

Read official docs thoroughly – Sub‑agent, Hooks, and Memory system in CodeBuddy are hard to use without them.

Prefer open‑source tools – Open Vibe Island suffices for monitoring; no need to pay for basic functionality.

Parallelism is the trend – Git Worktree + multi‑Agent workflow is becoming standard; not using it can cost an order of magnitude in efficiency.

Choose information sources by people, not topics – Tools like follow‑builders add value by following creators who actually ship products.

flow‑kit is worth watching – It relies only on Markdown, making it a durable best‑practice solution.

If you are using CodeBuddy or Claude Code for AI programming, this list covers the mainstream toolchain choices.

Follow me, James' growth diary, for ongoing practical insights to help you avoid detours in the AI era.