Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    User Flow and Sitemap Diagrams for Next.js and React Apps

    1

    Map real screens and navigation from Next.js, React Router, Vue, SvelteKit, Angular or Expo code into Mermaid sitemaps and journeys. Verdict: MAPPED, GAPS FOUND or BROKEN FLOWS.

    $7

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    User Flow and Sitemap Diagrams for Next.js and React Apps

    User Flow and Sitemap Diagrams for Next.js and React Apps

    Example session with this skill installed

    Map the screens in our Next.js app in apps/web (App Router) and draw the signup-to-first-project journey and checkout. Our PM wants to know which pages nobody can reach and whether anything is broken.

    • Read your context and instructions
    • Compiled the user flow
    • Generated the UI component

    User flows - apps/web @ 9c2f7d1 (Next.js 16.0, App Router)

    Verdict: BROKEN FLOWS Screens: 24 (3 modals) Edges: 61 (4 unconfirmed)
    Defaults used: journey threshold N = 7 Render test: passed 4/4

    Findings

    #SeverityFindingEvidenceFix
    1BROKENLink to /billing/plans, no such routeapp/(app)/settings/billing/page.tsx:42point to /pricing
    2BROKENRedirect loop for a user with onboarded=false and no teamproxy.ts:22 -> app/(app)/onboarding/layout.tsx:9send no-team users to /onboarding/team
    3GAPOrphan /changelog (no inbound link, not in emails)app/(marketing)/changelog/page.tsxadd footer link or delete
    4GAP/onboarding/step-2 has no back linkapp/(app)/onboarding/step-2/page.tsxadd back to step-1
    5GAPCheckout journey is 8 screens (N = 7)see journey tablemerge plan + seats steps

    Journey: signup to first project (6 screens)

    flowchart LR
      home(["/"]) -->|"1 · Start free"| signup["/signup"]
      signup -.->|"2 · submit ok"| verify["/verify-email"]
      subgraph gate["Signed in · proxy.ts:14"]
        s1["/onboarding/step-1"] -.->|"4 · next"| s2["/onboarding/step-2"]
        s2 -.->|"5 · finish"| first[["/projects/new"]]
      end
      verify ==>|"3 · verified"| s1
    
    StepFrom -> ToKindEvidence
    1/ -> /signuplinkapp/(marketing)/page.tsx:31
    2/signup -> /verify-emailpushapp/(auth)/signup/actions.ts:48
    3/verify-email -> /onboarding/step-1redirectapp/(auth)/verify-email/page.tsx:19

    Screen inventory (excerpt)

    RouteFileAuthInOut
    /projects/[id]app/(app)/projects/[id]/page.tsxsigned in57
    /changelogapp/(marketing)/changelog/page.tsxnone03

    Unconfirmed (not drawn)

    • app/(auth)/login/actions.ts:27 - redirect(searchParams.next), target from query string, not validated

    page.tsx

    TSX · React component

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Most teams do not have an accurate map of their own app. Figma flows drift from what ships, and an AI asked to "draw the user flow" from memory tends to invent tidy screens and links. This skill reads the router and the navigation calls in your code and draws only what it can trace to a file and line.

    What it does:

    • Detects the router and version from the lockfile: Next.js App Router and Pages Router, React Router (framework, data and declarative modes, including former Remix apps), Vue Router and Nuxt, SvelteKit, Angular, Expo Router and React Navigation
    • Builds a route inventory that handles route groups, dynamic and catch-all segments, layouts, parallel and intercepting routes (modals), lazy route configs and API routes
    • Extracts links, programmatic navigation, redirects and guards (proxy/middleware, loaders, beforeEach, canActivate, Nuxt middleware, Stack.Protected) with file:line evidence
    • Draws a Mermaid sitemap, 2-4 key journeys with auth-gated areas as subgraphs, and a screen inventory table with auth, params and inbound/outbound counts
    • Reports broken links, redirect loops, routes unreachable behind guards, orphan pages, dead ends, wizards without back or cancel, missing 404 routes and overlong journeys
    • Puts navigation it cannot resolve (targets from variables, APIs or ?next= parameters) in an Unconfirmed list instead of guessing
    • Includes a reference of router conventions, grep patterns, target resolution rules and Mermaid skeletons that were render-tested

    Who it's for: product managers, designers and QA engineers who need a current screen map, and developers onboarding to an unfamiliar frontend.

    Honesty note: the diagrams show what the code declares at one commit. Feature flags, A/B tests and CMS-driven menus can change real navigation, and the skill does not crawl the running app or read analytics. Client-side guards are reported, not treated as security.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Frontend & Web Apps.

    Maps what the code declares at one commit. Navigation from feature flags, CMS menus, API responses or query parameters is listed as Unconfirmed, not drawn. It does not run or crawl the app, read analytics, or verify server-side authorization. Framework conventions change between major versions, so it checks the repo's version first.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 29 days ago

    Software
    Git@mermaid-js/mermaid-cli
    Optional
    mermaid-cliNode.jsPython 3

    mermaid-cli runs on Node.js and test-renders diagrams. Python runs the link-graph check script.

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed29 days ago

    What's inside

    Frequently Asked Questions

    SKILL.md (the workflow, finding rules, verdict and output template) and references/router-conventions-and-patterns.md (router conventions per framework, grep patterns for links, redirects and guards, target resolution rules, finding exceptions, Mermaid skeletons and sources). That is the whole package.

    Any agent that reads SKILL.md skills, such as Claude Code, Codex CLI, Cursor, VS Code Copilot and Gemini CLI. The agent needs read access to the repo. Shell access for grep, Python and mermaid-cli improves results but is optional.

    Unzip the folder into your agent's skills directory (for example .claude/skills/ for Claude Code), open the frontend repo and ask for a sitemap, user flow or navigation audit. To render-test diagrams, install mermaid-cli or run it through npx.

    It reads code at one commit. It cannot see navigation driven by feature flags, CMS menus, API responses or query parameters, so those go to an Unconfirmed list. It does not run the app, crawl it or read analytics, and it cannot confirm that server-side authorization is correct.

    A plain prompt tends to produce a plausible flow. This skill makes the agent detect the router and version, inventory every route by framework rules, trace each link, redirect and guard to file:line, apply fixed rules for orphans, dead ends, broken links and loops, and give a repeatable verdict.

    While the skill captures all static and file-based routes, it cannot resolve URLs that are dynamically generated from external API responses or CMS-driven navigation menus.

    New in Frontend & Web Apps