AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Mobile Prototype

skill-xinvtech-mobile-design-os-mobile-prototype · by Xinvtech

Turn mobile UX, wireframe, UI, and platform contracts into a testable interaction prototype plan, a standalone browser review runner, or an implementation handoff. Use when Codex needs navigation transitions, gestures, motion, feedback, failure injection, runtime test cases, or an executable prototype scope.

No reviews yet
0 installs
16 views
0.0% view→install

Install

$ agentstack add skill-xinvtech-mobile-design-os-mobile-prototype

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-xinvtech-mobile-design-os-mobile-prototype)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
26d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Mobile Prototype? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Mobile Prototype

Make the core task observable in motion and runtime. Treat a prototype as a behavior test surface, not as a collection of clickable pictures.

Choose the prototype target

  • Behavior prototype: define screens, states, transitions, feedback, interruption, and test cases for a tool such as Figma, ProtoPie, or a native prototype.
  • Browser review runner: render a self-contained HTML design prototype from prototype.json, a UI Design contract, and an optional Design System contract. The default URL is a clean product preview; append ?inspect=1 for state continuity, contract edges, and deterministic failure review before production implementation.
  • Executable spike: create or hand off a narrow runnable prototype with real text, persistence boundaries, platform surfaces, and deterministic failure hooks.
  • Validation pass: execute the prototype test matrix and record evidence, limitations, and regressions.

Prefer mobile-wireframe before this Skill and mobile-design-to-code when the user wants framework code. Preserve upstream Screen, State, Flow, Requirement, Token, and Component IDs.

Define the runtime contract

  1. Select one critical task and one high-risk recovery path.
  2. Map each screen/state node to its entry, exit, action, system response, and durable result.
  3. Define animation, gesture, haptic, sound, and announcement behavior with a reduced-motion and nonvisual equivalent.
  4. Add interruption, background/termination, offline, duplicate submission, permission denial, and destructive recovery where applicable.
  5. Define deterministic failure hooks for persistence, network, indexing, permission, and system-surface cases instead of relying on luck.
  6. Keep prototype status honest: a plan or static render is not runtime evidence.

Read [references/prototype-rules.md](references/prototype-rules.md) and [references/prototype-schema.md](references/prototype-schema.md) before creating structured output.

Produce and validate

Create:

  • prototype-map.md using [assets/prototype-map-template.md](assets/prototype-map-template.md).
  • prototype.json using [assets/prototype-template.json](assets/prototype-template.json).
  • prototype-test-plan.md using [assets/prototype-test-plan-template.md](assets/prototype-test-plan-template.md).

Run python3 scripts/validate_prototype.py . Use the resulting not-tested records to drive implementation and device QA; do not convert them to pass without evidence.

When the prototype follows UI Design, also run python3 scripts/validate_ui_prototype_alignment.py --ui-design --prototype . This catches missing state nodes, unmapped explicit interactions, and high-risk states without a recovery hook.

For a clickable design-review surface, run the following after the contract checks. The supplied Design System must contain a locked visual direction copied from ui-design.json; an unlocked or legacy theme is a hard stop.

python3 scripts/render_prototype_html.py \
  --ui-design  \
  --design-system  \
  --prototype  \
  --output /prototype-runner/index.html
python3 scripts/validate_prototype_runner.py \
  --html /prototype-runner/index.html \
  --prototype 

The default runner URL must show only the product preview: no QA header, inspector, evidence badge, or debug phone chrome. The inspector remains available at ?inspect=1. The renderer must consume the locked visual fingerprint and semantic UI/Design System tokens for surface, text, action, border, typography, spacing, radius, color hierarchy, and module rhythm; do not reintroduce a generic card-stack, pill-button shell, or a legacy Margin theme. The runner must expose the product's primary task, state jump, outgoing contract edges, and deterministic failure switches for applicable persistence/index/permission/destructive paths in inspect mode. Its report remains runtimeEvidence: not-tested until browser, accessibility, interruption, and device evidence exists. A browser runner is a UI/prototype artifact, not production Flutter code.

Enforce the output order

  1. Prototype scope and evidence status.
  2. Screen/state nodes and transitions.
  3. Feedback, motion, gesture, system-surface, and interruption rules.
  4. Failure hooks and test matrix.
  5. Implementation handoff, evidence, and open risks.

Do not finish with a happy-path click-through that cannot show recovery, state continuity, accessibility pressure, or system behavior.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.