Claude Code cheat sheet: CLAUDE.md, Skills, MCP and the plan-implement-iterate loop
nikola_mr64990 · x · 2026-11-05
A practical Claude Code cheat sheet (with infographic) arguing the real advantage comes from knowing which command to use and structuring your project around it, not treating it as a chatbot.
- CLAUDE.md: project rulebook auto-generated via /init; focus on architecture, coding style, build commands, review checklists; keep under 150 lines
- Skills: reusable markdown prompt templates for code review, security scans, deployment checklists
- MCP: connect to local tools, databases, external APIs for live debugging
- Agents: spawn background sub-agents for parallel tasks like writing unit tests
- Pro tips: /context for usage, Hooks for auto-lint, /plan mode for safer exploration, /compact at 70% context
- Ideal workflow: Plan → Implement → Test → Review → Compact → Iterate
More from coding & agent
- Graphify maps code into knowledge graphs to fix vector search's blind spot for coding agents — JeremyCMorgan · 2026-11-05
- Notion gives all your agents one shared brain for memory and skills — mhdfaran · 2026-11-05
- OpenAI takes GPT-Live outbound telephony API to GA with SIP trunk support — gdb · 2026-11-05
- ReproEvalCard: a reporting standard for reproducible LLM pipeline evaluation — kalyan_kpl · 2026-11-05
- Harvard's Gary King Team Launches Certify: Agent-Automated Replication Certification for Journals — RexDouglass · 2026-11-05
- RETROGRADE benchmark round 0: 14 frontier models blind-tested, obfuscation hurts every one — amplifiedamp · 2026-11-05