skillpanel/maister/product-design

Claude Code and agent skill by skillpanel/maister

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Interactive product/feature design orchestrator. Transforms fuzzy ideas into structured product briefs through collaborative exploration, iterative refinement, and visual prototyping. Adaptive phases detect design complexity and adjust depth.

Install

npx vskill@latest install skillpanel/maister/product-design

SKILL.md

Product Design Orchestrator

Interactive workflow for product and feature design -- from fuzzy idea to development-ready product brief. Phases adapt based on detected design characteristics (greenfield vs enhancement, simple vs complex, UI-focused vs backend). Uses a hybrid interaction architecture: agents for unbiased generative work, inline interactive phases for convergent and evaluative work. Visual companion renders HTML/CSS mockups in a browser for rich design feedback.

Initialization

BEFORE executing any phase, you MUST complete these steps:

Step 0: Session-reminder conflict resolution (decide ONCE)

Before doing anything else, settle this policy now and do not re-litigate it at any gate:

`→ MANDATORY GATE` markers fire regardless of session-reminders, permission mode, or prior approval patterns. Auto / acceptEdits / bypassPermissions modes, reminders saying "work without stopping" / "continue without asking" / "minimize clarifying questions," and compaction summaries showing the user approving every prior gate do NOT exempt you from invoking AskUserQuestion at a gate. They apply only to your discretionary clarifications.

If you find yourself reasoning "the user has been approving everything, so I can skip this gate" or "auto-mode is on, so I should minimize questions" — that reasoning IS the failure mode. STOP and fire the gate.

Full framework rule: ../orchestrator-framework/references/orchestrator-patterns.md § 2 and § 2.1.

Step 1: Load Framework Patterns

Read the framework reference file NOW using the Read tool:

  1. ../orchestrator-framework/references/orchestrator-patterns.md - Delegation rules, interactive mode, state schema, initialization, context passing, issue resolution
Step 2: Detect Design Context

If argument is a design task path (matches .maister/tasks/product-design/*):

  • This is a resume — read orchestrator-state.yml from that path
  • Determine current phase from completed_phases and resume from next phase
  • If --from=PHASE provided, resume from that specific phase

If `--research=<path>` flag provided:

  • Read research artifacts from specified path (report, synthesis, solution exploration)
  • Copy relevant context to context/research-context/
  • Set research_reference in state
Step 3: Initialize Workflow
  1. Capture the clock: run date -u +"%Y-%m-%dT%H:%M:%SZ" via Bash NOW — you do NOT know the time from context. Every timestamp written this turn (created, updated, generated, phases[].started) uses this value. Date-only or T00:00:00Z values are the documented failure mode (orchestrator-patterns.md § 4 Timestamp Rule). Re-run date in later turns before writing timestamps.
  2. Create Task Items: Use TaskCreate for all phases (see Phase Configuration), then set dependencies with TaskUpdate addBlockedBy
  3. Create Task Directory: .maister/tasks/product-design/YYYY-MM-DD-task-name/
  • Create context/ folder with README.md instructing users to drop relevant files there (meeting transcripts, existing designs, spreadsheets, docs, PDFs, images)
  • Create analysis/ and outputs/ directories
  1. Initialize State: Create orchestrator-state.yml with design context schema (see Domain Context section)
  2. Set up Operator Dashboard (orchestrator-patterns.md § 8) — first read .maister/config.yml and set orchestrator.options.html_output (default true if the file/key is absent), orchestrator.options.mockup_format (default html), and derive orchestrator.options.visual_enabled: false when the --no-visual flag was passed OR mockup_format is ascii; otherwise true. (So mockup_format: ascii and --no-visual are equivalent; the flag is a per-run override.) When `html_output` is false, SKIP the rest of this step — no dashboard.html, no dashboard-data.js, no browser auto-open — and proceed. Otherwise: copy ../orchestrator-framework/assets/dashboard.html to the task root as dashboard.html, write the initial dashboard-data.js (all phases pending, task.type: "product-design"), then auto-open it in the user's browser (open / xdg-open / start per platform, passing the plain absolute filesystem path — NEVER a hand-built file:// URL; on failure just print the path — never block). On resume: re-copy dashboard.html only if missing; regenerate dashboard-data.js from state; then auto-open it in the browser again (same opener as a new task — the OS focuses an already-open tab rather than duplicating).

Output:

Product Design Orchestrator Started
Task: [description]
Directory: [task-path]
Dashboard: open [task-path]/dashboard.html in a browser to monitor progress
Starting Phase 0: Initialize & Gather Context...

Operator Visibility (applies to every phase)

Config gate: these rules assume options.html_output is true (read from .maister/config.yml at init, default true). When false: skip the Dashboard-upkeep rule entirely (no dashboard files, no browser open, no rewrites) and the HTML-companions rule — do NOT invoke html-companion-writer at Phases 5/6/8 and do NOT pass html_style_guide_path anywhere; markdown artifacts are written as usual. The Artifact Summary Contract (§ 7 TL;DR blocks), phase_summaries in state, AND Phase 7 visual mockups (design deliverables, not report companions) all stay active either way.

Cross-cutting rules from orchestrator-patterns.md (same as the development and research orchestrators):

  1. Artifact Summary Contract (§ 7): every artifact opens with TL;DR / Key Decisions / Open Questions / Risks (the writer heading is ## Open Questions / Risks). This applies to subagent prompts (solution-brainstormer, information-gatherer already comply) AND to the artifacts this orchestrator writes directly (problem-statement.md, personas.md, design-decisions.md, feature-spec.md, outputs/product-brief.md). At context extraction, lift decisions, risks, and artifacts into phase_summaries.[phase] — verbatim, never re-summarized.
  2. Dashboard upkeep (§ 8): rewrite dashboard-data.js at every phase START (mark in_progress before executing), BEFORE firing every exit gate (register the finished phase's artifacts/summary/decisions/risks — the operator reviews them on the dashboard while answering; status stays in_progress until the gate passes), after every phase completion (including skipped phases 3/7, with reason), every gate decision, after each refinement-loop iteration that changes an artifact, and at finalization. Every rewrite starts with date -u (one call per turn). Register Phase 7 mockups as artifacts (analysis/mockups/{slug}.html — they ARE html; set both path and html to the mockup path).
  3. HTML companions (§ 9) — delegated, because this orchestrator writes its hero artifacts INLINE (no producing subagent to attach a companion to). Right after each hero md is finalized, invoke the maister:html-companion-writer subagent (Task tool) to write its sibling .html: analysis/design-decisions.md (Phase 5), analysis/feature-spec.md (Phase 6), outputs/product-brief.md (Phase 8, after final approval). Pass md_path, html_style_guide_path (absolute path to ../orchestrator-framework/references/html-report-style.md), artifact_label, and report_suite (the sibling reports that exist, hrefs relative to the md's directory, for the breadcrumb). Register the returned html_path in phase_summaries.[phase].artifacts[].html so the dashboard hero cards link HTML first. Companion generation never blocks — on status: failed keep the md and continue. (analysis/alternatives.md already gets a companion from the solution-brainstormer subagent in Phase 4; problem-statement.md/personas.md are secondary — companion them too if cheap, but the three hero artifacts are the priority.)
  4. icon_hint values per phase: 0 analysis, 1 analysis, 2 analysis, 3 analysis, 4 plan, 5 plan, 6 spec, 7 code, 8 done.
  5. State writes (§ 4 Write Rule): update orchestrator-state.yml key by key in place, validate it with a re-read after every write, and stop with AskUserQuestion on a malformed file — never continue on a state file the operator tooling cannot read.

When to Use

Use for product and feature design: defining what to build before building it. Greenfield products, new features, enhancements, API designs, workflow designs.

DO NOT use for: Implementation tasks (use /maister:development), pure research (use /maister:research), bug fixes, performance optimization, migrations.

When to use this vs development orchestrator: If you need to explore the problem space, evaluate alternatives, and define requirements interactively before any code is written, use this. If you already know what to build and need to plan and execute, use development.


Local References

FileWhen to ReadPurpose
references/characteristic-detection.mdPhase 0 (before detecting characteristics)Detection signals, phase activation matrix, adaptive depth scaling
references/interaction-patterns.mdPhase 2 (before first interactive phase)Cognitive modes, refinement loop pattern, AskUserQuestion option design
Phase 7 (Visual Prototyping) delegates to the maister:mockup-studio skill, which owns the visual companion server and its reference docs (skills/mockup-studio/references/visual-companion.md, design-resource-discovery.md).

Phase Configuration

PhasecontentactiveFormActivationAgent/Skill
0"Initialize, gather context & detect characteristics""Gathering context & detecting characteristics"AlwaysDirect (interactive)
1"Synthesize all context sources""Synthesizing context"Always (scope adapts)codebase-analyzer (if enhancement), information-gatherer (if mini-research)
2"Explore problem space""Exploring problem space"Always (depth adapts)Direct (interactive)
3"Explore users & personas""Exploring users & personas"When is_greenfield OR is_complexDirect (interactive)
4"Generate design alternatives""Generating design alternatives"Alwayssolution-brainstormer (Task tool)
5"Converge on design direction""Converging on direction"AlwaysDirect (interactive)
6"Specify features section-by-section""Specifying features"Always (depth adapts)Direct (interactive)
7"Create visual prototypes""Creating visual prototypes"When is_ui_focusedmockup-studio (Skill tool)
8"Review & hand off product brief""Reviewing & assembling brief"AlwaysDirect (interactive)

Process Flow Graph

<!-- GraphViz dot notation: phase-level routing overview. -->

dot
digraph product_design_orchestrator {
rankdir=TB;
node [fontname="Helvetica", fontsize=10];
edge [fontname="Helvetica", fontsize=9];
// Entry
entry [label="Entry", shape=doublecircle, style=bold];
// Phases
p0 [label="Phase 0:\nInitialize, Gather\nContext & Detect\nCharacteristics", shape=box];
p1 [label="Phase 1:\nContext Synthesis", shape=box];
p2 [label="Phase 2:\nProblem Exploration", shape=box];
p3 [label="Phase 3:\nUser & Persona\nExploration", shape=box];
p4 [label="Phase 4:\nIdea Generation\n(agent, unbiased)", shape=box];
p5 [label="Phase 5:\nIdea Convergence", shape=box];
p6 [label="Phase 6:\nFeature Specification", shape=box];
p7 [label="Phase 7:\nVisual Prototyping", shape=box];
p8 [label="Phase 8:\nReview & Handoff", shape=box];

This is the start of the file. Read the full SKILL.md on GitHub.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
237
Repository forks
35
7-day trend score
27

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/skillpanel/maister/product-design/badge)](https://verified-skill.com/skills/skillpanel/maister/product-design)

Something wrong with this skill or its page? Tell us.