Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    User Profile, Media Upload & Onboarding Builder

    1

    Build, audit, and fix production user profile editors, cropped + cache-busted avatar/cover uploads, @username uniqueness, self-hiding completeness checklists, and first-run onboarding wizards — includes an offline static scanner for React, Next.js & Supabase.

    Free

    Security scanned
    User Profile, Media Upload & Onboarding Builder

    User Profile, Media Upload & Onboarding Builder

    Example session with this skill installed

    Audit our Next.js + Supabase /profile and /profile/[id] flows using user-profile-setup, fix any stale-avatar, privacy, or unscoped mutation bugs you find, and add a self-hiding profile completeness card so new users finish setting up their profile.

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

    I have completed the audit of your profile and onboarding flows. I scanned 8 files across app/profile/ and app/onboarding/ using the static analyzer.

    Profile & Onboarding Audit — Project-X

    Stack: Next.js + Supabase + shadcn/ui · Files scanned: 8
    Findings: 3 (High: 1, Medium: 1, Low: 1)

    1. [HIGH] MEDIA_CACHE_STALE: Missing Cache-Busting — app/profile/avatar-upload.tsx:42

    • What happens: The UI updates the src attribute to the same URL immediately after upload without a query token.

    User / security impact: The browser continues to show the old avatar from its local cache, leading users to believe the upload failed.

    Recommended fix: Append ?t=${Date.now()} to the public URL when updating the local state after a successful storage upsert.

    2. [MEDIUM] STATE_ROLLBACK_MISSING: No Cancel Snapshot — app/profile/edit-form.tsx:12

    • What happens: The handleCancel function simply closes the modal without reverting the local form state.

    User / security impact: If a user edits a field, cancels, and re-opens the editor, the stale unsaved changes persist in the UI.

    Recommended fix: Implement a snapshot ref on mount and call setForm(snapshot.current) within the cancel handler.

    3. [LOW] INPUT_TYPE_DATE_INVALID: Raw Date Format — app/profile/experience-list.tsx:88

    • What happens: The month input sends YYYY-MM strings directly to a Postgres date column.

    User / security impact: Potential for silent insertion failures or malformed data in environments strictly expecting YYYY-MM-DD.

    • Recommended fix: Normalize values to YYYY-MM-01 before sending the mutation.

    Next steps

    • Apply the high-priority fix for avatar cache-busting.
    • Refactor the profile editor to use the snapshot-rollback pattern.
    • Verify the fixes by re-running the audit_profile_flow.py script.

    avatar-upload.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

    Shipping a user profile and onboarding flow looks simple until real users hit the edge cases: re-uploaded avatars still show the old cached photo, cancelling a crop dialog breaks re-selecting the same image, @username collisions throw generic 500 toasts, sequential queries stall /profile load times, and select('*') on public /profile/[id] pages leaks internal account columns in DevTools.

    User Profile, Media Upload & Onboarding Builder (user-profile-setup) gives your coding agent a battle-tested, production-extracted workflow plus an offline static analyzer to build, audit, and refine user profile systems without guesswork or dependency bloat.

    Four Operating Modes

    1. BUILD — Scaffolds or extends /profile editors, multi-step /onboarding wizards, avatar/cover crop + upload pipelines, @username validation, sub-collection managers (skills, tags, work history, pinned/hidden endorsements), and self-hiding completeness cards using your repo's existing UI kit.
    2. AUDIT — Runs the bundled read-only scripts/audit_profile_flow.py static scanner across your codebase, mapping concrete bugs (PF001–PF008) to exact file:line locations ranked by severity before touching a single line of code.
    3. REFINE — Applies minimal, surgical patches with snapshot/cancel rollback, optimistic UI updates, and defense-in-depth owner scoping.
    4. VERIFY — Re-runs the static scanner alongside your project's typecheck/test suite (tsc --noEmit, Vitest/Jest) and reports a clean before/after delta.

    What the Bundled Static Scanner Catches (audit_profile_flow.py)

    RuleWhat It Detects
    PF001Profile or sub-table update() / delete() calls missing explicit .eq('user_id', user.id) owner scoping
    PF002Deterministic upsert: true avatar/cover uploads missing ?t= URL cache-busting (stale image bug)
    PF003Hidden <input type="file"> handlers missing e.target.value = '' reset (broken same-file re-select)
    PF004Media upload handlers missing client-side file.size guards before network upload
    PF005Sequential await waterfalls inside profile/onboarding loaders instead of Promise.all
    PF006Icon-only camera/remove/pin <button> controls missing aria-label (WCAG 4.1.2)
    PF007@username / handle updates missing Postgres 23505 / Prisma P2002 duplicate-handle feedback
    PF008Public /profile/[id] or /u/[handle] routes running select('*') on profiles

    What's Inside the Package

    • SKILL.md — Stack-adaptive intake + 4-mode workflow (BUILD, AUDIT, REFINE, VERIFY) and strict owner-vs-public route boundaries
    • scripts/audit_profile_flow.py — Pure Python 3 stdlib, 100% offline, read-only CLI scanner (--format text and --format json)
    • references/state-and-save-patterns.md — Snapshot rollback, parallel hydration, handle normalization, <input type="month"> SQL date formatting, and pinned-first sub-lists
    • references/media-upload-and-cropping.md — Hidden file input wiring, 1:1 & banner crop contracts, cache-busting, and storage bucket policies
    • references/onboarding-and-completeness.md — onboarding_completed mount guard, searchable country/tag pickers, auto-hiding completeness math, and verification eligibility deep-linking
    • references/public-vs-owner-security.md — Public column allowlists, hidden testimonial filtering, owner-scoped writes, and idempotent RLS templates
    • examples/real-session-audit-and-patch.md — Complete worked audit-to-verification session

    Supported Stacks

    Works out of the box with React, Next.js (App & Pages Router), Remix, SvelteKit, and Vue paired with Supabase, Postgres, Prisma, or Drizzle and Tailwind / shadcn / Base UI. Zero external Python packages or network access required.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.

    Does not handle auth provider setup, billing/subscriptions, or global search algorithms. Static scanner requires Python 3.8+. Optimized for Supabase, Prisma, and Drizzle stacks.

    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

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    • Free with an Agensi account

    Software
    tscNode.js

    Permissions

    Environment Variables
    Network Access
    Write Files
    Terminal / Shell
    Listed19 days ago

    What's inside

    Frequently Asked Questions

    This skill streamlines the creation and auditing of user profiles, handling complex production requirements like avatar cropping, optimistic UI updates, and secure owner-only route protections that generic AI prompts often overlook.

    It is optimized for Next.js, React, and TypeScript environments, with native support for Supabase, Prisma, and Drizzle, ensuring it integrates seamlessly with modern full-stack architectures.

    The skill performs a static scan of your repository to identify UX waterfalls, accessibility gaps, and security risks in your existing profile routes, providing surgical patches to fix them.

    While it excels at UI/UX and data management for profiles, it does not handle authentication protocols like OAuth, password resets, or third-party billing integrations like Stripe.

    The offline static scanner requires Python 3.8+ to run, but the core workflow orchestration is designed to work within any standard AI agent environment that supports the provided tags.

    New in Prompt & Skill Engineering

    Free