Compound Dev
Make Claude Code sessions compound — every session builds on the last instead of starting from scratch. Implements Karpathy's LLM Wiki pattern for software engineering. Use this skill when starting a new project, onboarding to an existing codebase, or any time the user says "set up my project", "add project memory", "create CLAUDE.md", "help me structure this project", "I keep re-explaining things to Claude", "my sessions don't remember anything", "ingest this document", or "run a health check". Also trigger when the user asks about session efficiency, token usage, or wants to improve their Cl
Install
npx skills add brijoobopanna/ClaudeSkills --skill compound-devgit clone --depth 1 https://github.com/brijoobopanna/ClaudeSkills.git /tmp/ClaudeSkills
mkdir -p ~/.claude/skills
cp -r /tmp/ClaudeSkills/compound-dev ~/.claude/skills/compound-devPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# compound-dev Make every Claude Code session build on the last one. Implements Karpathy's [LLM Wiki](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f) pattern for software engineering: raw sources are compiled into a persistent, interlinked wiki maintained entirely by the LLM. You steer. The LLM writes. Every session compounds. --- ## The core idea Most Claude Code sessions work like RAG — the LLM rediscovers your project from scratch every time. You re-explain the architecture, re-debate decisions, re-discover patterns. Nothing accumulates. compound-dev is different. The LLM **incrementally builds and maintains a persistent wiki** — structured, interlinked markdown files that capture decisions, patterns, gotchas, entities, people, and concepts. When you start a session, the LLM reads the wiki and already knows. When you end a session, new knowledge integrates into the existing wiki — updating entity pages, noting contradictions, strengthening the synthesis. **The wiki is a persistent, compounding artifact.** The cross-references are already there. The contradictions have been flagged. The synthesis reflects everything you've built. The wiki gets richer with every session. This is not a wiki for the sake of having a wiki. It's a **context engineering system.** You're building the exact input your LLM needs to do useful work. Every meeting note, every linked decision, every filed artifact improves the
Details
- Skill name
compound-dev- Repository
- brijoobopanna/ClaudeSkills
- Path
compound-dev/SKILL.md- License
- MIT
- Repo updated
- 5 April 2026
FAQ
What does the Compound Dev skill do?
Make Claude Code sessions compound — every session builds on the last instead of starting from scratch. Implements Karpathy's LLM Wiki pattern for software engineering. Use this skill when starting a new project, onboarding to an existing codebase, or any time the user says "set up my project", "add project memory", "create CLAUDE.md", "help me structure this project", "I keep re-explaining things to Claude", "my sessions don't remember anything", "ingest this document", or "run a health check". Also trigger when the user asks about session efficiency, token usage, or wants to improve their Cl
How do I install the Compound Dev skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add brijoobopanna/ClaudeSkills --skill compound-dev. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Compound Dev popular?
It lives in brijoobopanna/ClaudeSkills, which has 9 GitHub stars and 0 forks.
Related skills
Visual Brief
9brijoobopanna/ClaudeSkills
Create publication-quality technical infographics from any source — tweets, articles, architectures, or ideas. Use this skill whenever the user wants to visually break down a technical concept, map a system architecture, turn a tweet or thread into a shareable diagram, create editorial-style pipeline or flow visualizations, illustrate feedback loops or compounding mechanisms, or produce visuals for a newsletter or blog. Trigger on phrases like "visualize this", "diagram this", "break this down visually", "map this architecture", "make an infographic", "turn this tweet into a visual", "show me
Coordinator Orchestrator
9kadaliao/claude-code-skills-collection
多 Agent 协调者模式 — 你是指挥官,不是执行者。调度 Worker 做研究/实现/验证,你负责综合和决策。适用于:(1) 复杂任务需要多步骤并行 (2) 需要研究→综合→实现→验证的工作流 (3) bobooo 说"用 CC 做"时的编排模式。灵感源自 Claude Code Coordinator Mode。
Adversarial Verification
9kadaliao/claude-code-skills-collection
对抗性验证技能 — 验证代码变更/部署/任务交付物的正确性。不是确认"它能工作",而是尝试"把它搞坏"。用于:(1) 完成非平凡任务后自验 (2) 审查 PR/MR (3) 部署后验证 (4) bobooo 说"验证一下"。灵感源自 Claude Code Verification Agent 设计。
Event Modeling
9noahseger/agent-skills
Use to model a system's behavior as domain events, extract an event model from existing code, design event-driven architectures, or document workflows as timelines of commands, events, and state views
Multi Agent Verify
9leyuan0602-glitch/cc-claude-codex
Multi-agent parallel verification. Spawns 3 independent agents (OpenCode CLI, Codex CLI, Claude Task subagent) in separate git worktrees to review, test, and fix code. Synthesizes results and applies final fixes on the original branch.
Gemini Nano Banana 2
9tamtom/image-generation-skills
Generate and edit images using Google's Gemini Nano Banana 2 (gemini-3.1-flash-image-preview). Use when the user asks to generate, create, edit, modify, change, alter, or update images using Gemini. Also use when user references an existing image file and asks to modify it in any way. Supports text-to-image generation, image editing, multi-image compositing (up to 14 reference images), style transfer, Google Web Search and Image Search grounding for real-time data, high-resolution output up to 4K, controllable thinking levels, 14 aspect ratios, and advanced text rendering. DO NOT read the imag
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.