Skip to content
Menu

Vercel Plugin for AI Coding Agents

Last updated April 1, 2026

The Vercel plugin turns any supported AI coding agent into a Vercel expert. It pre-loads agents with a relational knowledge graph of the entire Vercel ecosystem and automatically injects the right guidance at the right time based on what you're working on.

The plugin includes four main components that work together to enhance your AI coding agent:

ComponentDescription
Ecosystem graphA relational knowledge graph covering every Vercel product, library, CLI, API, and service, with decision matrices and cross-product workflows
38 skillsDeep-dive guidance for specific Vercel products, injected automatically when relevant
3 specialist agentsPurpose-built agents for deployment, performance optimization, and AI architecture
5 slash commandsQuick actions for deploying, managing environment variables, bootstrapping projects, and more
ToolStatus
Claude CodeSupported
CursorSupported
OpenAI CodexComing soon
  • One of the supported AI coding tools listed above
  • Node.js 18+

Install the plugin with a single command:

npx plugins add vercel/vercel-plugin

The plugin activates automatically after installation. There are no additional setup steps or commands to learn.

After installing, the plugin detects what you're working on from your tool calls, file paths, and project configuration, then injects the right expertise at the right time. You use your AI agent as you normally would and the plugin handles the rest.

The plugin uses lifecycle hooks that run during your session:

  • Session start context injection: Injects the ecosystem graph into every session so the agent has baseline knowledge of the full Vercel platform
  • Session start repo profiler: Scans config files and dependencies to prepare skill matching for a faster first tool call
  • Pre-tool-use skill injection: When the agent reads, edits, or writes files, the plugin matches file paths and commands against skill patterns and injects relevant guidance
  • Pre-write/edit validation: Before the agent writes or edits a file, the plugin catches deprecated patterns like sunset packages, old API names, and renamed files

Skills are the core of the plugin's expertise. Each skill covers a specific Vercel product or feature in depth. The plugin injects up to three skills per tool call, prioritized by relevance, and deduplicates across the session so the same skill is never injected twice.

For example:

  • Editing a next.config.ts file triggers the nextjs skill
  • Running vercel deploy triggers the deployments-cicd skill
  • Working with useChat or streamText triggers the ai-sdk skill

The plugin includes 38 skills covering the full Vercel ecosystem:

SkillCovers
agent-browserBrowser automation CLI: dev server verification, page interaction, screenshots, form filling
agent-browser-verifyAutomated dev-server verification: visual gut-check on page load, console errors, key UI elements
ai-elementsPre-built React components for AI interfaces: chat UIs, tool call rendering, streaming responses
ai-gatewayUnified model API, provider routing, failover, cost tracking, 100+ models
ai-sdkAI SDK v6: text/object generation, streaming, tool calling, agents, MCP, providers, embeddings
authAuthentication integrations: Clerk, Descope, Auth0 setup for Next.js with Marketplace provisioning
bootstrapProject bootstrapping orchestrator: linking, env provisioning, db setup, first-run commands
chat-sdkMulti-platform chat bots: Slack, Telegram, Teams, Discord, Google Chat, GitHub, Linear
cron-jobsVercel Cron Jobs configuration, scheduling, and best practices
deployments-cicdDeployment and CI/CD — deploy, promote, rollback, --prebuilt, CI workflow files
env-varsEnvironment variable management: .env files, vercel env commands, OIDC tokens
investigation-modeOrchestrated debugging: runtime logs, workflow status, browser verify, deploy/env triage
json-renderAI chat response rendering — UIMessage parts, tool call displays, streaming states
marketplaceIntegration discovery, installation, auto-provisioned env vars, unified billing
next-cache-componentsNext.js 16 Cache Components: PPR, use cache, cacheLife, cacheTag, updateTag
next-forgeProduction SaaS monorepo starter: Turborepo, Clerk, Prisma/Neon, Stripe, shadcn/ui
next-upgradeNext.js version upgrades: codemods, migration guides, dependency updates
nextjsApp Router, Server Components, Server Actions, Cache Components, routing, rendering strategies
observabilityWeb Analytics, Speed Insights, runtime logs, Log Drains, OpenTelemetry, monitoring
react-best-practicesReact/Next.js performance optimization: 64 rules across 8 categories
routing-middlewareRequest interception before cache, rewrites, redirects, personalization with Edge/Node.js/Bun runtimes
runtime-cacheEphemeral per-region key-value cache, tag-based invalidation, shared across Functions/Middleware/Builds
shadcnshadcn/ui: CLI, component installation, custom registries, theming, Tailwind CSS integration
sign-in-with-vercelOAuth 2.0/OIDC identity provider, user authentication via Vercel accounts
turbopackNext.js bundler, HMR, configuration, Turbopack vs Webpack
turborepoMonorepo orchestration, caching, remote caching, --affected, pruned subsets
v0-devAI code generation, agentic intelligence, GitHub integration
vercel-agentAI-powered code review, incident investigation, SDK installation, PR analysis
vercel-apiVercel MCP Server and REST API: projects, deployments, env vars, domains, logs
vercel-cliAll CLI commands: deploy, env, dev, domains, cache management, MCP integration, marketplace
vercel-flagsFeature flags, Flags Explorer, gradual rollouts, A/B testing, provider adapters
vercel-functionsServerless, Edge, Fluid Compute, streaming, Cron Jobs, configuration
vercel-queuesDurable event streaming, topics, consumer groups, retries, delayed delivery
vercel-sandboxEphemeral Firecracker microVMs for running untrusted/AI-generated code safely
vercel-servicesMultiple services in one project: monorepo backends + frontends on the same domain
vercel-storageBlob, Edge Config, Neon Postgres, Upstash Redis, migration from sunset packages
verificationFull-story verification: infers user story, verifies end-to-end browser → API → data → response
workflowWorkflow SDK: durable execution, DurableAgent, steps, Worlds, pause/resume

The plugin includes three specialist agents that you can invoke for focused tasks:

AgentExpertise
deployment-expertCI/CD pipelines, deploy strategies, troubleshooting, environment variables
performance-optimizerCore Web Vitals, rendering strategies, caching, asset optimization
ai-architectAI application design, model selection, streaming architecture, MCP integration

Invoke commands directly within your AI coding agent:

CommandPurpose
/vercel-plugin:bootstrapBootstrap a project with linking, env provisioning, and db setup
/vercel-plugin:deployDeploy to Vercel (preview or production)
/vercel-plugin:envManage environment variables (list, pull, add, remove, diff)
/vercel-plugin:statusView project status, recent deployments, and environment overview
/vercel-plugin:marketplaceDiscover and install Vercel Marketplace integrations

To deploy to production, pass prod as an argument: /vercel-plugin:deploy prod

If the plugin isn't injecting skills when expected, enable debug logging by setting the VERCEL_PLUGIN_LOG_LEVEL environment variable:

export VERCEL_PLUGIN_LOG_LEVEL=debug

Available log levels:

LevelDescription
offNo logging (default)
summaryHigh-level injection summaries
debugDetailed matching and dedup information
traceFull pipeline traces with timing breakdowns

You can also use the built-in doctor command to diagnose issues:

npx vercel-plugin doctor

This validates manifest parity, checks hook timeouts, and verifies dedup health.

If a skill gives incorrect advice or injection doesn't fire when expected, file an issue on GitHub. Include:

  • What you were building
  • What the plugin injected (or didn't). Enable debug logs with VERCEL_PLUGIN_LOG_LEVEL=debug
  • What was wrong about it

Was this helpful?

supported.