--- title: Prompts description: Build an app, extend an agent, or improve what you have with prompts you can use in your coding agent. url: "https://vercel.com/kb/prompts" related_resources: "[Blog](/blog)": "Read engineering stories and practical guides for building on Vercel." "[Changelog](/changelog)": "Catch up on new features, platform improvements, and product releases." "[Templates](/templates)": "Build your next app with a starter you can customize and deploy." --- # Prompts Build an app, extend an agent, or improve what you have with prompts you can use in your coding agent. - [Add a tool to an existing agent](/kb/prompts/add-agent-tool): Extend: Extend your agent with one focused tool, including validated inputs, server-side permissions, and tests for service failures. - [Add feature flags you can test in preview deployments](/kb/prompts/add-preview-feature-flags): Extend: Put one feature behind a safe default and add Vercel Toolbar overrides so your team can review both states in previews. - [Add human approval before your agent takes action](/kb/prompts/add-human-approval-to-agent-actions): Extend: Add a review step to an existing AI SDK tool, with explicit approval, server authorization, and protection against repeated actions. - [Add regression tests for a critical user flow](/kb/prompts/add-critical-flow-regression-tests): Improve: Protect a key Next.js user journey with isolated tests for visible outcomes, failure recovery, and relevant access boundaries. - [Add resumable background workflows to your app](/kb/prompts/add-resumable-background-workflows): Extend: Move an existing background job into Workflow steps, with bounded retries, visible progress, and protection against duplicate effects. - [Add verifiable source citations to your AI answers](/kb/prompts/add-citations-to-ai-answers): Extend: Extend an existing AI SDK app with source citations that readers can inspect, plus a clear response when evidence is missing. - [Build a document extraction app with structured output](/kb/prompts/build-document-extraction-app): Build: Turn document text into validated fields with the AI SDK, including missing-value handling and a reviewable JSON result. - [Build a private file upload app with Vercel Blob](/kb/prompts/build-private-file-upload-app): Build: Build an owner-only file library with private Vercel Blob storage, authenticated downloads, and checks for cross-user access. - [Build a scheduled data sync that recovers from failures](/kb/prompts/build-scheduled-data-sync): Build: Create a bounded data sync with authenticated scheduling, durable checkpoints, and tests for retries, duplicate runs, and recovery. - [Build a searchable knowledge base for your content](/kb/prompts/build-searchable-knowledge-base): Build: Create a Next.js knowledge base with readable article pages, shareable search, and results drawn from your full content collection. - [Build a support triage agent with eve](/kb/prompts/build-eve-support-triage-agent): Build: Create a local eve agent that summarizes support requests, identifies missing information, and suggests a next step for review. - [Build a webhook event viewer for local debugging](/kb/prompts/build-webhook-event-viewer): Build: Inspect verified Vercel webhook events in a local viewer with redacted records, duplicate detection, and reproducible fixtures. - [Build an AI chat app with streaming responses](/kb/prompts/build-ai-chat-app): Build: Build a chat interface with the AI SDK and AI Gateway, including streamed responses, cancellation, and recoverable errors. - [Connect your eve agent to a read-only MCP service](/kb/prompts/add-eve-mcp-connection): Extend: Add one approved MCP service to eve with reviewed read operations, explicit account credentials, and checks for denied or expired access. - [Delegate document reviews to an eve subagent](/kb/prompts/add-eve-review-subagent): Extend: Add a document-review specialist to your eve agent with bounded context, explicit results, and handling for failed or incomplete reviews. - [Evaluate an agent before changing its prompt](/kb/prompts/evaluate-agent-prompt-changes): Improve: Build a repeatable evaluation for an AI SDK agent, compare prompt versions, and surface failures hidden by an average score. - [Find and fix a slow PostgreSQL query in your app](/kb/prompts/fix-slow-database-query): Improve: Investigate one slow PostgreSQL query with representative test data, preserve its results, and compare execution plans. - [Improve your app’s accessibility](/kb/prompts/improve-app-accessibility): Improve: Find and fix reproducible accessibility barriers in a key user journey, combining automated checks with keyboard and focus testing. - [Keep AI responses available with model fallbacks](/kb/prompts/add-ai-gateway-model-fallbacks): Extend: Configure AI Gateway fallbacks for an existing generation flow, checking model compatibility, error handling, and retry boundaries. - [Keep customer access isolated in your eve agent](/kb/prompts/isolate-eve-customer-access): Improve: Review an eve app’s session and service boundaries, then test that one customer cannot read another customer’s conversations or data. - [Protect a public API with verified rate limits](/kb/prompts/protect-public-api-from-abuse): Improve: Add server-enforced limits to one API route, test rejection and recovery, and document the Firewall configuration it needs. - [Reduce the JavaScript your app sends to browsers](/kb/prompts/reduce-client-javascript): Improve: Trim client JavaScript on one Next.js route, measure the difference, and preserve the interactions your users need. - [Teach your eve agent a release-readiness checklist](/kb/prompts/add-eve-release-checklist-skill): Extend: Add an eve skill that reviews release evidence, identifies blockers, and separates passing checks from work that remains unverified. ## Keep building Take your prompt further with engineering guides, platform updates, and templates for your next app or agent. - [Blog](/blog): Read engineering stories and practical guides for building on Vercel. - [Changelog](/changelog): Catch up on new features, platform improvements, and product releases. - [Templates](/templates): Build your next app with a starter you can customize and deploy. ## Reference Architecture Choose a stack and work through the decisions behind it, with curated documentation, guides, templates, and courses to help you build. - [Agent stack](/kb/architecture/agents): Build and operate agents with models, tools, and durable workflows. - [Web applications](/kb/architecture/web): Choose rendering, data, authentication, and caching approaches. - [Backends](/kb/architecture/backends): Choose a runtime, connect services, and run background work. [Browse all stacks](/kb/architecture)