By bpainter
Eleven specialist skills for shipping on Vercel — deploy pipeline (Production / Preview / Development), Fluid Compute and runtimes, the Edge Network and CDN, Vercel-managed storage (Blob, KV, Postgres, Edge Config), observability (Speed Insights, Web Analytics, Logs, Drains), security (WAF, BotID, Bot Management, Deployment Protection), the Vercel AI SDK and AI Gateway, v0 design-to-code, the Sandbox + Agent + Workflows agentic runtime, and the Vercel REST API. Adjacent to the software-engineering plugin's Next.js skills — that plugin owns the framework, this plugin owns the platform.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Use Vercel's three agentic-infrastructure primitives together — Sandbox (sandboxed code execution from agent tools), Vercel Agent (hosted agentic runtime with built-in state and tools), and Workflows (durable execution for multi-step processes that survive restarts, retries, and human approvals). Use this skill any time agents need to execute generated code safely, when long-running multi-step processes need durability, when "this needs to run for hours / wait for a human / retry on failure" comes up, or when an existing serverless-function approach is straining under workflow complexity. Trigger on any agentic-infrastructure question on Vercel.
Use the Vercel AI Gateway as the single edge layer in front of every AI provider — multi-provider routing across OpenAI / Anthropic / Google / Mistral / xAI / etc., fallback chains for provider outages, aggregated observability (token usage, cost, latency, errors), cost guardrails, and BYO-keys vs. Vercel-billed aggregation. Use this skill any time AI features are being built or extended on Vercel, when "we keep hitting OpenAI rate limits" comes up, when multiple model providers are in play, when a cost dashboard is needed, or when a fallback strategy is being designed. Trigger on any AI routing or AI cost question.
Build AI features with the Vercel AI SDK (ai-sdk.dev) — streaming text and structured output, model adapters across providers, tool calling, RSC integrations, useChat / useCompletion / useObject hooks, and the wiring patterns that pair the SDK cleanly with the AI Gateway, Fluid Compute, and Vercel-managed storage. Use this skill any time an AI feature is being built on Next.js, when the team is asking "how do we ship a chat UI", when streaming or tool calling is in scope, when RSC + AI is the question, or when an existing AI integration needs to be standardized. Trigger any time AI SDK or `ai` package usage is implicated.
Make the Vercel Edge Network and CDN do what you want — cache layers and headers, ISR / PPR behavior on Vercel's runtime, on-demand revalidation, image optimization metering, bandwidth cost guardrails, and when to bypass to an external CDN (Cloudflare, Bunny). Use this skill when caching is misbehaving, when bandwidth or transform costs are spiking, when "why is this stale" or "why is this slow" questions land, or when a launch is imminent and the cache strategy needs review. Trigger on any "edge" or "CDN" question on Vercel.
Stand up and operate the Vercel deploy pipeline — project setup, the Production / Preview / Development environment trio, custom environments, domains and aliases, Deployment Protection (SSO / password / Bypass Tokens), build configuration, rollback via re-aliasing, and the deploy-hook patterns that drive CI-from-non-git sources. Use this skill when a new Vercel project is being created, when "set up our preview deployments" comes up, when staging deploys need a stable URL, when domain configuration is in question, when a deploy fails and the cause isn't obvious, or when rollback is on the table. Trigger on any "how do we ship to Vercel" question.
Uses power tools
Uses Bash, Write, or Edit tools
The Composable DXP plugin marketplace for Claude. 18 plugins, 182 skills, packaged for Claude Code (CLI), Claude Desktop (Cowork mode), and drag-and-drop install.
Marketplace name: composable-dxp (the value in .claude-plugin/marketplace.json)
Repo: https://github.com/bpainter/composable-dxp-claude-marketplace (public, personal)
# Cowork (Git URL — Cowork requires Git, not local paths)
claude plugin marketplace add https://github.com/bpainter/composable-dxp-claude-marketplace.git
claude plugin install <plugin>@composable-dxp
# Claude Code (also accepts the local path)
claude plugin marketplace add "/Users/bermon.painter/Library/CloudStorage/OneDrive-Slalom/Slalom Second Brain/80_Skills_and_Agents"
claude plugin install <plugin>@composable-dxp
For full install paths (Cowork, Claude Code, drag-and-drop), see INSTALL.md.
80_Skills_and_Agents/
├── README.md # this file (conventions and structure)
├── INSTALL.md # how to install the marketplace and plugins
├── .claude-plugin/
│ └── marketplace.json # marketplace manifest — lists every plugin
├── Memory/ # cross-cutting working memory (CLAUDE.md, people, glossary, etc.)
│ # NOT a plugin — read by Claude every session
├── Plugins/ # built .plugin files (build artifacts)
│ └── <plugin>.plugin
└── <plugin-name>/ # one folder per plugin SOURCE (e.g., obsidian/, claude/)
└── ... (see "Plugin layout" below)
The marketplace itself is a single Claude plugin marketplace. Tools that consume marketplaces (Claude Code CLI, Claude Desktop via Cowork) read .claude-plugin/marketplace.json to discover what's installable.
Memory/ is not a plugin. It's the user's working memory and shared knowledge base, read every session by Claude, and shared across all plugins.
Plugins/ holds built artifacts — the .plugin files (renamed ZIPs) that get dragged into Claude Desktop's Personal Plugins panel for install. The source of each plugin lives in its own folder at the marketplace root (e.g., obsidian/, claude/); the .plugin files in Plugins/ are derived from those sources via the build process described in INSTALL.md.
Every plugin under this marketplace follows this exact shape. Deviations break tooling.
<plugin-name>/ # plugin SOURCE (lives at marketplace root)
├── README.md # plugin overview for humans
├── .claude-plugin/
│ └── plugin.json # plugin manifest for Claude (only `name` is required)
├── references/ # plugin-wide shared references (optional)
│ └── <plugin>-foundations.md
├── skills/ # auto-discovered by Claude
│ ├── <plugin>-<skill-name>/ # one folder per skill
│ │ ├── SKILL.md # required — frontmatter (name, description) + body
│ │ └── references/ # skill-specific deep references (typical, 1–3 files)
│ │ └── <topic>.md
│ └── ... (more skills)
├── agents/ # auto-discovered by Claude (workers)
│ └── <plugin>-<agent-name>.md # single file per agent, Claude Code agent format
└── commands/ # slash commands (optional)
└── <command-name>.md
| Piece | Required? | Purpose |
|---|---|---|
README.md | recommended | Plugin overview, opinion, roadmap, written for humans browsing the repo |
.claude-plugin/plugin.json | required | Manifest Claude reads. Only name is strictly required; version, description, author, keywords, license are optional. Don't include empty optional fields (e.g., "homepage": "") — they fail manifest validation |
references/ | optional | Plugin-wide shared references at the plugin root. From a SKILL.md, cite as ../../references/<file>.md |
skills/ | optional | Skills (one folder per skill, each with a SKILL.md). Auto-discovered by Claude |
skills/<skill>/SKILL.md | required if folder exists | The skill itself — YAML frontmatter (name, description) + body |
skills/<skill>/references/ | optional | Skill-specific deep references (frameworks, checklists, examples). From the SKILL.md, cite as references/<file>.md |
agents/ | optional | Workers that act on user data — Claude Code agent file format (one file per agent) |
commands/ | optional | Slash commands — single Markdown files with description: frontmatter and a body that names the skill or workflow to invoke |
A plugin can have any combination of skills, agents, and commands. Empty folders are fine if you intend to fill them later.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub bpainter/composable-dxp-claude-marketplace --plugin vercelClaude tooling skills for Cowork and Claude Code. Two skills: claude-plugin-creator scaffolds new plugins (folder structure, manifest, .plugin build, Cowork install) with empirically-established validation rules; claude-orchestrator designs the orchestration layer for new Cowork projects (CLAUDE.md routing, multi-phase commands, agent teams). One slash command (/plugin-help).
Seven behavioral-economics specialists grounded in canonical sources — Kahneman's Thinking Fast and Slow, Thaler & Sunstein's Nudge, Ariely's Predictably Irrational, and Wendel's Designing for Behavior Change — plus Slalom's 6-stage Behavioral Design Model. Six ready-to-use templates: Behavioral Profile, Choice Architecture, Nudges/Biases, Intervention Design, Use-Case Scoring, Use-Case Validation.
Eleven innovation specialists grounded in canonical sources — Christensen's Innovator's Dilemma, Dyer/Gregersen/Christensen's Innovator's DNA, Furr & Dyer's Innovator's Method, Doblin's Ten Types, Ramanujam & Tacke's Monetizing Innovation, Innovation Lab Excellence, Kelley's Ten Faces, Catmull's Creativity Inc, Dyer/Furr/Lefrandt's Innovation Capital, Rogers's Driving Digital Strategy, Tushman/Anand's Why Digital Transformations Fail, and Kane et al's Technology Fallacy — plus HBR Must-Reads on Innovation and Design a Better Business. Skills cover strategy, portfolio governance, value engineering across horizons, JTBD discovery, lean validation, ten-types diagnosis, willingness-to-pay monetization, lab design, disruption analysis, creative leadership/political capital, and digital transformation. One agent (innovation-agent) sequences them. Eight slash commands (audit, strategy, portfolio, discover, validate, value-case, charter, disruption-test) provide high-leverage recurring operations. Six templates: Three Horizons canvas, Ambition Matrix, JTBD Interview Guide, Ten Types diagnostic, Value Engineering canvas, Innovation Charter.
Leadership coaching, people-management, hiring, meetings and cadences, goal-setting, transitions, and stoic perspective. Seven specialized skills under one orchestrating agent. The two original skills (executive coach for neurodivergent leaders, people-leader for everyday craft) are joined by skills for hiring, meetings & cadences, onboarding & transitions, goals & OKRs, and stoic perspective. Each skill carries deep reference files distilled from canonical leadership texts — Hire With Your Head, The First 90 Days, Radical Focus, Drive, Good to Great, The Coaching Habit, Never Split the Difference, Meeting Design, The Effective Manager, Managing Humans, The Manager's Path, A Guide to the Good Life, How F*cked Up Is Your Management, Peopleware, Scaling Teams, Performance Appraisal Phrase Book, Exactly What to Say, How We Talk, Radical Candor, and 7 Habits.
Six brand specialists: strategy (positioning, narrative, voice, audience), naming (five-category framework, six-step process), identity-system (logo direction, color, type, photography, iconography), voice-tone (attributes with not-X pairings, tone register matrix, filler-words bans), guidelines-composer (9-chapter deliverable, sized to brand context), and audit (drift detection across identity, voice, surface application, governance). Outputs feed the design plugin and marketing plugin.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Binary reverse engineering, malware analysis, firmware security, and software protection research for authorized security research, CTF competitions, and defensive security
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.