Use when you want to see everything L.L.O.Y.D. can do — full capability overview including all skills by category, MCP servers, plugins, and agents.
Logical Learning & Optimization Yield Director Named after Lloyd Lee, Ari Gold's elite assistant in Entourage. Overqualified, unflappable, always delivers.
Run these checks to show the live state of the config, then present the full summary.
# Skill count
ls ~/.claude/skills/ | wc -l
# MCP servers
cat ~/.claude/settings.json
Report everything below in a clean, grouped format.
| Plugin | Version | What it does |
|---|---|---|
| superpowers |
| v5.0.5 |
| Full workflow framework — TDD, planning, subagents, debugging, code review, git worktrees |
| oh-my-claudecode | v4.9.1 | UI and theme enhancements for Claude Code |
Always-on, available in every project session.
| Server | What it does | Requires |
|---|---|---|
| context7 | Injects live, up-to-date library docs into context | Nothing |
| playwright | Browser automation and UI testing via natural language | Nothing |
| github | Read repos, open issues, create PRs, search code | GITHUB_PERSONAL_ACCESS_TOKEN in ~/.bashrc |
| ruflo | Spawn Claude agent swarms, coordinate parallel tasks | ruflo installed globally |
| context-mode | Context switching between modes | Nothing |
Verify all are connected: run /doctor in Claude Code.
Pre-built role definitions in agents/. Pass their path to the Agent tool when dispatching subagents.
| Agent | Role |
|---|---|
| agents/code-reader.md | Read-only exploration — finding files, reading source, searching patterns. Use before implementation. |
| agents/verifier.md | Validation — runs tests, builds, linting. Use after implementation to confirm correctness. |
| agents/searcher.md | Research — web searches, documentation lookup. Use when gathering external context. |
These are role definitions, not skills — dispatch via the Agent tool, not the Skill tool.
Invoke any skill via the Skill tool. Type / in Claude Code to browse all invocable skills.
| Skill | When to use |
|---|---|
superpowers:brainstorming | Before any creative work — features, architecture, naming |
superpowers:writing-plans | Before touching code on a multi-step task |
superpowers:executing-plans | Executing a written plan in a separate session |
superpowers:subagent-driven-development | Executing a plan with parallel/independent tasks in current session |
superpowers:test-driven-development | Before writing any implementation code |
superpowers:systematic-debugging | When hitting a bug or failing test |
superpowers:requesting-code-review | After completing a feature or before merging |
superpowers:receiving-code-review | When acting on review feedback |
superpowers:verification-before-completion | Before claiming any task is done |
superpowers:finishing-a-development-branch | When implementation is complete and ready to ship |
superpowers:dispatching-parallel-agents | When 2+ independent tasks can run simultaneously |
superpowers:using-git-worktrees | When feature work needs isolation from main workspace |
superpowers:writing-skills | When creating or editing a skill |
| Skill | When to use |
|---|---|
sync-dotfiles | Sync LLOYD to a new machine or pull latest updates |
update-claude-md | Auto-generate or update a project's CLAUDE.md |
skill-creator | Guided workflow for building a new skill |
| Skill | When to use |
|---|---|
last30days | Deep research across Reddit, X, HN, YouTube, Polymarket, Brave (last 30 days) |
| Skill | When to use |
|---|---|
copywriting | Writing landing page / homepage copy |
copy-editing | Improving existing copy |
content-strategy | Planning content (blog, SEO topics) |
seo-audit | SEO audit or technical SEO issues |
schema-markup | Structured data / rich results |
ai-seo | AI search optimization (ChatGPT, Perplexity) |
programmatic-seo | Programmatic content at scale |
site-architecture | Site structure / URL architecture |
paid-ads | Paid ads (Google, Meta, LinkedIn) |
ab-test-setup | A/B testing with statistical rigor |
signup-flow-cro | Signup flow optimization |
onboarding-cro | Onboarding flow optimization |
page-cro | Landing page conversion optimization |
popup-cro | Popup / modal optimization |
form-cro | Form optimization |
paywall-upgrade-cro | In-app paywall / upgrade screen optimization |
pricing-strategy | Pricing / packaging decisions |
cold-email | Cold outbound email sequences |
email-sequence | Drip / nurture email sequences |
social-content | Social media content |
ad-creative | Ad creative (headlines, visuals, copy) |
launch-strategy | Product launch planning |
churn-prevention | SaaS churn analysis / retention |
referral-program | Referral or affiliate programs |
competitor-alternatives | Competitor comparison pages |
analytics-tracking | Analytics / event tracking setup |
revops | Revenue ops / lead lifecycle |
sales-enablement | Sales collateral / battle cards |
marketing-psychology | Behavioral psychology applied to UX |
marketing-ideas | Brainstorming marketing tactics |
lead-magnets | Lead magnet planning and creation |
free-tool-strategy | Free tool for marketing / SEO |
product-marketing-context | Product marketing context document |
| Skill | When to use |
|---|---|
frontend-design | Building production web UI |
ui-ux-pro-max | UI/UX decisions, color, typography |
web-design-guidelines | Reviewing UI for accessibility / best practices |
web-artifacts-builder | Multi-component HTML artifacts |
algorithmic-art | Generative / algorithmic art (p5.js) |
canvas-design | Static poster / visual art (.png, .pdf) |
theme-factory | Applying a visual theme to an artifact |
brand-guidelines | Applying Anthropic brand colors and typography |
slack-gif-creator | Animated GIFs optimized for Slack |
remotion-best-practices | Video creation in React (Remotion) |
| Skill | When to use |
|---|---|
docx | Creating or editing Word documents |
pdf | PDF manipulation (merge, split, extract) |
pptx | Creating PowerPoint presentations |
xlsx | Spreadsheet creation or analysis |
| Skill | When to use |
|---|---|
mcp-builder | Building an MCP server |
webapp-testing | Testing a local web app with Playwright |
agent-sandboxes | Running code in a secure cloud sandbox |
internal-comms | Internal docs, status updates, newsletters |
doc-coauthoring | Co-authoring a spec or proposal |
LLOYD maintains persistent memory at ~/.claude/projects/*/memory/. Each project gets its own memory directory. Memories are recalled automatically at session start when relevant.
Types stored: user profile, feedback/corrections, project context, external references.
echo "Skills: $(ls ~/.claude/skills/ | wc -l)"
echo "MCP servers: $(cat ~/.claude/settings.json | python -c "import sys,json; print(len(json.load(sys.stdin).get('mcpServers', {})))")"