agentsclimarketplace

04 inspiration

Skill andrefabricio/prodotypor/files/.claude/skills/04-inspiration

An AI-assisted product development pipeline for Claude Code. Go from a product idea to Figma screens in 8 structured steps, with specialized AI agents handling product management, design research, and visual construction.

Install
npx -y skills add andrefabricio/prodotypor --skill 04-inspiration

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Step 4 of 8 — UI inspirational research. Performs structured design audits of reference applications to inform visual direction. Can run in PARALLEL with step 03 (PRD creation). Sub-commands: [continue] resume, [status] check progress, [update] force status refresh.

SKILL.md

5.5 KB, as published. Nobody here has run it

/04-inspiration — UI Inspirational Research

Research and document UI/UX design patterns from reference applications.

Sub-command Router

Parse $ARGUMENTS:

  • Contains [status] → Jump to Status Report section
  • Contains [update] → Jump to Force Status Update section
  • Contains [continue] or empty → Jump to Execute section
  • Other text → Treat as app name to research

Prerequisites

Hard dependencies (flexible — this step can start early):

  1. At least ONE of these must be true:
    • Step 03 (PRD) status == completed — full feature context available
    • .impeccable.md exists — brand direction available for research lens
  2. If NEITHER exists, report: "Step 04 needs at least a brand direction (.impeccable.md) or completed PRDs to focus the research."

Parallel execution note: This step can run in PARALLEL with step 03. It does not require PRD output — inspiration research is independent competitive analysis. However, PRD context improves research relevance.

Soft dependencies:

  • PRDs (improve relevance of research — know what features need design inspiration)
  • Existing inspiration docs (avoid duplicate research)

Execute

Phase 1: Determine Research Scope

  1. Read .impeccable.md — extract aesthetic direction, anti-references, brand personality
  2. Read PRDs if available — note feature types that need design inspiration
  3. Check existing research in docs/Design/Inspirations/ — avoid duplicates
  4. If $ARGUMENTS names a specific app, target that app
  5. If no app specified, identify 2-3 candidates based on:
    • Similar emotional context or user sensitivity to {{PROJECT_NAME}}
    • Similar feature patterns (feeds, cards, timelines, circles, etc.)
    • Recognized design quality
  6. Present candidates to the user for approval (if discovering independently)

Phase 2: Delegate to Inspiration Researcher Agent

For each app to audit, spawn the inspiration-researcher agent:

Perform a comprehensive design audit of [app name].

Project context: {{PROJECT_NAME}} — [summary from CLAUDE.md] Brand direction: [from .impeccable.md] Feature scope: [from PRDs if available] Template: [content of templates/inspiration-research-template.md]

Requirements:

  • Professional-grade design audit (≥10KB)
  • All 6 sections from the template
  • Specific measurements (hex values, pixel sizes, font names)
  • Explicit relevance assessment for {{PROJECT_NAME}}
  • Save to docs/Design/Inspirations/[app-name]-review.md

Pipeline Context

Before delegating, read and summarize prior step outputs (if available):

  • Step 03 output (if completed): read PRD feature list for targeted research focus
  • Step .impeccable.md: brand direction for research lens
  • Include this summary in the agent prompt as "Pipeline Context"

Phase 3: Validate Output

  1. Verify at least 2 inspiration review files exist in docs/Design/Inspirations/
  2. Check each review has all 6 sections from the template
  3. Check each review is ≥10KB (sufficient depth)
  4. If validation passes → proceed to Phase 4
  5. If validation fails → proceed to Error Recovery

Error Recovery

If validation fails:

  1. Identify the specific failure (e.g., "review is only 5KB, need ≥10KB" or "missing Typography section")
  2. Feed the error + partial output back to the agent:

    "The inspiration review did not meet requirements: [specific failure]. Current output: [summary of what sections exist and their sizes]. Please expand: [specific sections that need more depth]."

  3. Re-validate after the retry
  4. If retry also fails, mark step as blocked (reason) in docs/workflow-status.md

Quality Gate (optional)

<!-- CUSTOMIZE: Define domain-specific quality checks for inspiration research -->

After structural validation, optionally assess quality:

  • Are color palettes documented with actual hex values (not vague descriptions)?
  • Are typography specs specific enough to inform a design document?
  • Score: pass / needs-revision

Phase 4: Human Review Gate

  1. Update docs/workflow-status.md step 04 status to needs-review
  2. Report research summary to the user
  3. Prompt: "Review the inspiration audits. Run /04-inspiration [continue] to approve, or provide feedback."
  4. Only [continue] after needs-review transitions to completed

Phase 5: Report

Step 04 — UI Inspiration Research: [status]

Reviews completed: [count]
  - [app-name]: [file size], [section count] sections
  - [app-name]: [file size], [section count] sections

Key patterns identified:
  Adopt: [brief list]
  Avoid: [brief list]

Next step: /05-design-doc (requires both step 03 + step 04 completed)

Status Report

  1. Read docs/workflow-status.md — extract step 04 section
  2. List review files in docs/Design/Inspirations/ with sizes
  3. Report completion (need ≥2 reviews)

Force Status Update

  1. Glob docs/Design/Inspirations/*.md — count reviews
  2. Check file sizes (≥10KB threshold)
  3. Update docs/workflow-status.md step 04

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.