Reusable agents, skills, standards, hooks, and MCP configurations for Claude Code. Drop this into your ~/.claude directory to get an opinionated, productivity-focused setup.
LLMs suffer from context rot — too much information degrades response quality. This configuration addresses that through agent delegation: specialized agents get their own clean contexts while an architect orchestrates from above. Each agent reports back, keeping both the top-level and agent contexts small and focused.
The top-level orchestration and the architect run on Opus, while focused task agents (developer, debugger, reviewer) run on cheaper models (Sonnet/Haiku) that work well for scoped work.
For a deeper overview of this pattern, read ClaudeLog: You Are the Main Thread.
Specialized personas delegated to by the /do orchestrator.
| Agent | Role | Model |
|---|---|---|
| architect | Solution design, ADRs, technical recommendations | Opus (default) |
| developer | Code implementation from specifications | Sonnet |
| debugger | Root cause analysis through systematic evidence gathering | Haiku |
| quality-reviewer | Security, performance, and concurrency review | Haiku |
| research | Multi-perspective research and synthesis | Haiku |
| technical-writer | Concise documentation for completed features | Sonnet |
Automated multi-step workflows invoked by trigger phrases or slash commands.
Linear Integration:
| Skill | Description |
|---|---|
| linear | Issue tracking CLI with semantic search and velocity analytics |
| start | Start a Linear issue — check blockers, assign, create branch, plan, execute |
| finish | Finish an issue — check requirements, commit/push, mark Ready For Release |
| triage | Analyze backlog for staleness, blockers, and priority suggestions |
| prd | Create agent-friendly tickets with PRDs and success criteria |
| cycle-plan | Plan cycles using velocity analytics and historical capacity |
| retro | Analyze completed cycles for retrospectives |
| deps | Visualize issue dependency chains and circular dependencies |
| link-deps | Discover and link related issues as dependencies |
Development Workflow:
| Skill | Description |
|---|---|
| pr-update | Generate PR titles and descriptions from actual code changes |
| dependency-updater | Orchestrate dependency updates with research and validation |
| deprecation-handler | Migrate deprecated APIs with safe patterns |
| semver-advisor | Classify version changes as MAJOR/MINOR/PATCH |
| react-component-generator | Generate React 19+ components with TypeScript |
External Skills (installed by update.sh):
| Skill | Source | Description |
|---|---|---|
| agent-browser | vercel-labs/agent-browser | Browser automation for AI agents |
| skill-creator | vercel-labs/agent-browser | Guide for creating new skills |
| vercel-react-best-practices | vercel-labs/agent-skills | React/Next.js performance optimization |
| vercel-composition-patterns | vercel-labs/agent-skills | React composition patterns that scale |
Universal rules governing agent behavior. See standards/README.md.
| Standard | Covers |
|---|---|
| agent-coordination | Parallel vs sequential execution patterns |
| git | Commit messages, destructive command blocking, multi-session safety |
| problem-solving | When to stop and ask vs proceed |
| technical-debt-prevention | No backups, no duplicates, delete aggressively |
| semver | Version classification and compatibility rules |
| version-aware-planning | Check actual versions before planning |
| deprecation-handling | Proactively update deprecated code |
| project-commands | Always use project-specific scripts |
Path-specific conventions applied automatically when editing matching files.
| Rule | Applies To |
|---|---|
| typescript | **/*.ts, **/*.tsx |
| react | **/*.tsx, **/*.jsx |
| markdown | **/*.md, **/*.mdx |
| package-manager | **/package.json, lockfiles |
Automatic quality checks that run without manual invocation.
| Hook | Trigger | What It Does |
|---|---|---|
| git-permissions | Before git commands | Blocks destructive operations (reset --hard, --force, clean -f) |
| lint-post-tool | After file edits | Runs Biome and markdownlint with auto-fix |
| typecheck | On stop | Runs tsc -b after TypeScript changes |
| Command | Description |
|---|---|
| /do | Plan execution — breaks work into phases, delegates to specialized agents, validates each step |
curl -fsSL https://claude.ai/install.sh | bashAssuming you already have a ~/.claude directory from using Claude Code, add this repo:
cd ~/.claude
git init
git remote add origin https://github.com/alienfast/claude.git
git fetch
git checkout -b main origin/main
git pullNote: This supplements your user directory with reusable configurations — it does not overwrite personal settings or data. Always check before committing to ensure no local user data is included, and adjust
.gitignoreaccordingly.
~/.claude/update.shThis installs:
- Vercel agent-browser and skill-creator
- Vercel React best practices and composition patterns
- Linear CLI and all Linear skills
- Runs markdown linting
See mcpServers.md for available MCP server configurations. The current approach favors skills over MCP servers for context efficiency — most MCP servers have been removed in favor of CLI tools and skills.
Primary entry point for complex, multi-step work:
/do I want to update Traefik. Search traefik documents, compare the version
we are currently on, and what we might need to change to be up to date.
Implement the changes.
/do this code was originally written for react 16. While some files have been
updated for react 19, I want you to take a look at a comprehensive review of
all react code, and implement the best practices for react 19.
Skills activate by trigger phrase or slash command:
/prdfor interactive linear issues creation with dependencies, subtasks etc or use existing research like/prd @doc/my-research.md/start PL-123— Start working on a Linear issue/finish— Complete the current issueupdate the PR— Generate/update PR title and descriptionupdate dependencies— Run the dependency updater workflow/triage— Analyze and prioritize backlog
No manual intervention needed — hooks run behind the scenes:
- Destructive git commands are blocked before execution
- Biome and markdownlint run after every file edit
- TypeScript type checking runs when Claude stops
Create a directory in skills/ with a SKILL.md containing YAML frontmatter. See skills/README.md for the full guide.
Add a markdown file to standards/. It will be referenced by agents automatically. See standards/README.md.
Add a markdown file to rules/ and register the glob pattern in CLAUDE.md under "Path-Specific Rules."