claude-code-best-practice/README.md

claude-code-best-practice

your best ai assistant with the best practice

CONCEPTS

Note: Custom slash commands have been merged into skills. Files in .claude/commands/ still work, but skills (.claude/skills/) are recommended as they support additional features like supporting files, invocation control, and subagent execution.

• Skills - Reusable knowledge, workflows, and slash commands that Claude can load on-demand or you invoke with /skill-name
• Subagents - Isolated execution contexts that run their own loops and return summarized results
• Memory - Persistent context via CLAUDE.md files and @path imports that Claude sees every session
• Rules - Modular topic-specific instructions in .claude/rules/*.md with optional path-scoping via frontmatter
• Hooks - Deterministic scripts that run outside the agentic loop on specific events
• MCP Servers - Model Context Protocol connections to external tools, databases, and APIs
• Plugins - Distributable packages that bundle skills, subagents, hooks, and MCP servers
• Marketplaces - Host and discover plugin collections
• Settings - Hierarchical configuration system for Claude Code behavior
• Permissions - Fine-grained access control for tools and operations

Extension Overview: See Extend Claude Code for when to use each feature and how they layer together.

LESSON LEARNED

My Experience

■ Workflows

• Claude.md should be 100 lines only.
• Have feature specific subagents (extra context) with skills (progressive disclosure).
• /memory, /rules, constitution.md does not guarentee anything
• do manual /compact at max 50%
• subtasks should be so small that it can be completed in less than 50% session of context window
• always start with plan mode
• use human pass gated todo list workflow
• vanilla cc is better than 3rd party workflows

■ Utilities

• wispr-flow for voice prompting (10x productivity)
• voice hooks for claude feedback
• status line for glancing compact
• use git worktress
• use /config dont ask permission mode instead of dangerously—skip--permissions

■ Debugging

• /doctor
• always ask claude to run the terminal (you want to see logs of) as a background task for better debugging
• use mcp (claude in chrome, playwright, chrome dev tool) to let claude see chrome console logs on its own

Context Engineering

• Humanlayer - Writing a good Claude.Md

Workflows

• Ralph plugin with sandbox
• Human Layer RPI - Research Plan Implement
• AgentOs - 2026 its overkill (Brian Casel)
• Github Speckit
• GSD - Get Shit Done
• OpenSpec OPSX
• Superpower

LIBRARIES

• Claude Code Tips
• Awesome Claude Code

KEYWORDS

Claude Keywords

• [btw] start prompt with btw to let the current task executing in background
• [claude --dangerously-skip-permissions] "defaultMode": "bypassPermissions"
• deprecated* [ultrathink] Triggers extended thinking with up to ~32K thinking tokens for a single request.

Community

• Agentic Workflow
• AI Slop
• Context Engineering
• Context Rot
• Dumb Zone
• Hallucination
• One Shot
• Rate Limit Jail / Cooldown
• Slot Machine Method = Save state → let Claude run → revert if bad results → try again
• The Holy Trinity = Skills + Agents + Hooks working together
• Token Burn
• Vibe Coding

CLAUDE CODE FEATURES INSPIRATION

• Claude Code Tasks - inspired by beats Inspiration
• Ralph Plugin

DOCS

docs/AGENTS.md - Agent Orchestration Best Practices

Learn how to properly orchestrate multiple agents, avoid common pitfalls when sub-agents aren't invoking, and implement sequential workflows using the Task tool.

docs/PROMPTS.md - Invocation Patterns Reference

Quick reference tables showing how to invoke agents and commands from different contexts (CLI, other agents, other commands).

docs/WEATHER.md - Weather System Flow Documentation

Complete system architecture and flow diagram for the weather data fetching and transformation workflow, demonstrating real-world agent and command orchestration.

IDE

Cursor vs. Windsurf vs. VS Code

Claude Code Replaced Cursor for Me… Here’s Why