Claude Code is my primary coding environment. Over the last year it has grown from a chat wrapper into a full operating layer — planning, building, testing, shipping, and running my startup all happen through it. This is a clean tour of what I have wired into it and what each piece does.
Plugins
everything-claude-code
A massive curated library of coding standards, patterns, reviewers, build-error resolvers, and domain skills. It gives Claude opinionated references for any language, framework, or stack, plus workflow pieces like security reviews, test-driven development, and agent engineering. Think of it as a shared brain that raises the floor on everything Claude writes.
superpowers
Process discipline. Instead of letting Claude jump straight to code, this plugin enforces brainstorm → plan → test → implement → verify. Skills include brainstorming, writing-plans, executing-plans, test-driven-development, systematic-debugging, verification-before-completion, and using-git-worktrees. Rigid by design — you follow it, you do not negotiate with it.
claude-mem
Persistent memory across sessions. Claude normally forgets everything between conversations; claude-mem maintains a database of decisions, discoveries, and lessons so future sessions pick up where the last one left off. /mem-search finds prior context, /timeline-report narrates project journeys over weeks.
ui-ux-pro-max
Design intelligence for web and mobile. Seven skills covering design systems, UI styling, brand, banners, and slides. Opinionated enough that output does not look like another generic Tailwind template.
skill-creator
Meta-plugin for building and editing your own skills. When you notice yourself pasting the same prompt across projects, this is the tool that turns it into a reusable skill.
caveman
Output compression mode. Drops articles, filler, pleasantries, and hedging from Claude's text — cuts roughly 75% of output tokens without losing technical substance. Still writes normal English for code, commits, and security warnings.
gstack
Browsing, QA, and shipping toolkit. Provides /browse (headless Chromium for dogfooding), /qa (systematic app testing), /review → /ship → /land-and-deploy (full pipeline), /canary (post-deploy watch), /benchmark (perf regression), and /investigate (four-phase root-cause debugging). Installed globally so every project gets it.
GSD — Get Shit Done
A project-structure system. Eighty-plus commands for running real work with a paper trail. The core loop is /gsd-new-project → /gsd-discuss-phase → /gsd-plan-phase → /gsd-execute-phase → /gsd-verify-work → /gsd-ship. Each phase gets its own directory, plan, atomic commits, and verification checklist.
MCP servers
MCP servers are bridges out of Claude into real systems. Each one gives Claude a set of tools that hit an external API.
Productivity
- Notion — search, fetch, create, and update workspace pages and databases.
- Gmail — authenticate and work with mail.
- Google Calendar — list, create, update, delete events, suggest meeting times.
- Google Drive — auth for file access.
- Linear — full CRUD across issues, projects, cycles, documents, milestones.
Docs and research
- Context7 — live documentation lookup for any library, framework, SDK, or CLI. Overrides stale training data with current docs.
- LiveKit-Docs — LiveKit Agents docs, SDK code search, changelog.
- Exa — neural search across web, code, and companies.
Dev tools
- GitHub — full repo / issue / PR / branch / file operations.
- Playwright — browser automation (click, navigate, screenshot, fill forms, evaluate JS). Backs the QA and design-review skills.
- Memory — in-session entity-relation knowledge graph.
- Sequential-thinking — structured multi-step reasoning.
Persistent memory
- claude-mem mcp-search — queries the persistent memory database. Exposes
search,smart_search,get_observations,timeline, and corpus management.
Skills — auto-invoked (daily drivers)
These load by default and Claude matches them to context on its own.
Writing code
When editing Python, TypeScript, or Next.js, pattern skills like python-patterns, python-testing, pytorch-patterns, nextjs-turbopack, frontend-patterns, backend-patterns, api-design, postgres-patterns, database-migrations, docker-patterns, deployment-patterns, git-workflow, hexagonal-architecture, architecture-decision-records, and coding-standards quietly shape the output.
Testing
tdd-workflow— enforces write-tests-first.e2e-testing— Playwright patterns for end-to-end coverage.ai-regression-testing— regression strategies for AI-assisted development.
AI and agent engineering
claude-api— Anthropic SDK patterns with prompt caching baked in.mcp-server-patterns— build new MCP servers with the Node/TS SDK.agent-harness-construction— design agent action spaces and tool definitions.agent-evalandeval-harness— measure whether changes to an agent actually improve it.
Security
security-reviewandsecurity-scanas defaults.
Research
documentation-lookupvia Context7.exa-searchfor semantic web research.deep-researchfor multi-source investigations.
Product
product-lens— validate the "why" before building.product-capability— translate PRD intent into concrete capabilities.accessibility— WCAG-aware design and audit.
Ops (always on)
github-ops— repository operations and automation.terminal-ops— evidence-first repo execution workflow.
Workflow (gstack + GSD core)
/browse,/qa,/review,/ship,/land-and-deploy,/canary,/investigate,/devex-review,/learn,/freeze,/unfreeze,/careful,/guard/context-save,/context-restore- GSD core loop:
/gsd-plan-phase,/gsd-execute-phase,/gsd-verify-work,/gsd-ship,/gsd-debug,/gsd-code-review,/gsd-progress,/gsd-next,/gsd-health,/gsd-stats,/gsd-discuss-phase,/gsd-spec-phase,/gsd-research-phase
Process + memory
- superpowers (all): brainstorming, writing-plans, executing-plans, TDD, systematic-debugging, verification-before-completion, requesting-code-review, receiving-code-review, dispatching-parallel-agents, subagent-driven-development, finishing-a-development-branch, writing-skills, using-git-worktrees
- claude-mem (all):
/mem-search,/smart-explore,/make-plan,/do,/timeline-report,/knowledge-agent,/version-bump - ui-ux-pro-max (all 7 design skills)
- caveman modes
Skills — on-demand (manual /invoke)
These are loaded but hidden from Claude's auto-match to keep the session context clean. I call them by name when the situation shows up.
Business
/investor-materials— pitch decks, one-pagers, investor memos./investor-outreach— cold emails, warm intros, follow-ups./market-research— competitive analysis, investor due diligence./lead-intelligence— AI-native lead intelligence pipeline./content-engine— platform-native content systems for X, LinkedIn, TikTok./brand-voice— source-derived writing style profile./article-writing— articles, guides, tutorials, newsletters./seo— technical SEO audits and remediation./ui-demo— polished Playwright-recorded product demo videos./frontend-slides— animation-rich HTML presentations.
Ops
/email-ops— evidence-first mailbox triage, drafting, send verification./messages-ops— live multi-channel messaging./google-workspace-ops— Drive, Docs, Sheets, Slides as one unified workflow./project-flow-ops— GitHub + Linear triage and flow./knowledge-ops— KB management, ingestion, sync, retrieval./research-ops— evidence-first current-state research workflow./automation-audit-ops— automation inventory and overlap audit./customer-billing-ops— subscriptions, invoicing, refunds./finance-billing-ops— revenue, pricing, team billing./unified-notifications-ops— GitHub + channels + email notifications as one flow./ecc-tools-cost-audit— ECC burn and billing audit.
AI and agent specialty
/cost-aware-llm-pipeline— cost optimization patterns for LLM API usage./agent-introspection-debugging— structured self-debug for agent failures./enterprise-agent-ops— long-lived agent observability and security./ai-first-engineering— operating model for teams where agents generate most code./agentic-engineering— eval-first execution mindset./api-connector-builder— build new API connectors or providers./dashboard-builder— build monitoring dashboards that answer real operator questions./data-scraper-agent— fully automated AI-powered data collection agent.
Meta and config
/configure-ecc— Everything Claude Code installer./skill-stocktake— audit Claude skills and commands for quality./skill-comply— visualize skill compliance./hookify-rules— create hooks from conversation patterns./prompt-optimizer— analyze and rewrite raw prompts./token-budget-advisor— pick response length informed by cost./strategic-compact— suggest compaction at logical breaks./workspace-surface-audit— audit active repo, MCP, plugins, connectors./repo-scan— cross-stack source asset audit./codebase-onboarding— generate structured onboarding doc./code-tour— persona-targeted step-by-step code tours./regex-vs-llm-structured-text— decision framework regex vs LLM.
Planning extras
/search-first— research-before-coding workflow./iterative-retrieval— progressively refine retrieval./blueprint— one-line objective into construction plan./opensource-pipeline— fork, sanitize, package private repo for open source.
Plan reviews
/plan-ceo-review— founder / CEO lens./plan-eng-review— eng manager lens./plan-design-review— designer lens./plan-devex-review— developer experience lens./plan-tune— adjust review sensitivity./autoplan— runs all four reviews in sequence.
Design
/design-consultation— end-to-end design brief./design-shotgun— multiple AI design variants./design-html— production-quality HTML finalization./design-review— designer-eye QA.
Gstack extras
/cso— Chief Security Officer infra-first mode./retro— weekly engineering retrospective from commits./document-release— post-ship documentation update./office-hours— YC-style advisory mode./codex— OpenAI Codex CLI wrapper for cross-model review./pair-agent— pair a remote AI agent with your browser./gstack,/gstack-upgrade,/health,/benchmark-models— stack management./setup-browser-cookies,/setup-deploy,/connect-chrome,/open-gstack-browser— one-time setup helpers.
GSD extras
- Phase-specific:
/gsd-ui-phase,/gsd-ai-integration-phase,/gsd-sketch,/gsd-sketch-wrap-up,/gsd-spike,/gsd-spike-wrap-up. - Idea capture:
/gsd-note,/gsd-plant-seed,/gsd-add-backlog,/gsd-review-backlog,/gsd-add-todo,/gsd-check-todos. - Doc management:
/gsd-import,/gsd-ingest-docs,/gsd-docs-update. - Triage + branch:
/gsd-inbox,/gsd-pr-branch. - Context handoff:
/gsd-pause-work,/gsd-resume-work,/gsd-thread,/gsd-session-report,/gsd-extract_learnings. - Milestone lifecycle:
/gsd-new-milestone,/gsd-milestone-summary,/gsd-complete-milestone,/gsd-cleanup,/gsd-plan-milestone-gaps. - Phase structure:
/gsd-add-phase,/gsd-insert-phase,/gsd-remove-phase,/gsd-analyze-dependencies,/gsd-list-phase-assumptions,/gsd-undo,/gsd-add-tests. - Multi-repo workspaces:
/gsd-new-workspace,/gsd-list-workspaces,/gsd-remove-workspace,/gsd-workstreams. - Auto + quick:
/gsd-autonomous,/gsd-quick,/gsd-do,/gsd-explore. - Meta:
/gsd-profile-user,/gsd-set-profile,/gsd-manager,/gsd-reapply-patches,/gsd-review.
Security extras
/security-bounty-hunter— hunt exploitable issues before someone else does.
Modes I run
Caveman
On by default in new sessions. Compresses Claude's output. Off with stop caveman or normal mode.
Auto-memory
File-based memory system at ~/.claude/projects/.../memory/. Per-project user, feedback, project, and reference notes. Separate from claude-mem.
CLAUDE.md router
Global instructions file at ~/.claude/CLAUDE.md. Tables map situations to the right skill to call. Keeps hidden / manual-only skills discoverable without loading them into Claude's auto-listing every session.
Why this matters
Software engineering is not getting slower. AI-native tools are becoming the primary interface, and the quality of your setup is becoming a real differentiator. The list above is not a flex — it is an attempt to build a personal operating system where the machine handles the work I should not be doing by hand.
The goal is not to install everything. The goal is to install exactly the things that earn their keep and route the rest out of sight until they are needed.