prd-interactive-review
This skill should be used when the user asks to "make the PRD an HTML", "render this PRD as HTML to review", "build an interactive PRD", "let me vet the PRD as HTML", "turn this spec into something I can click through", or wants to review and approve a PRD/spec by inspecting an immersive HTML rendering instead of reading raw markdown. It produces a self-contained, tabbed, single-file HTML review experience with SVG flow diagrams — not a flat colorized dump of the .md.
What this skill does
# PRD Interactive Review
## Purpose
Render a finished or near-final PRD (or comparable spec) as a self-contained, interactive HTML artifact whose single job is to make **vetting and approving** the document faster and more immersive than reading the source markdown. The deliverable exploits HTML's interactive nature — tabbed navigation, SVG flow diagrams with clickable detail, progressive disclosure, and color used as a signal — so a reviewer grasps the critical points and signs off with minimal effort.
This is **not** a markdown-to-HTML converter. A long scrolling page that merely adds color "misses the point of HTML." Optimize for comprehension-before-sign-off, not for reproducing every line.
## Core principles
1. **Purpose first.** The artifact exists to be inspected and approved. Be faithful to the source on critical points only — not 1:1. Keep the source `.md` canonical for complete text.
2. **Single file.** Inline all CSS and JS. No CDN or external assets. Must work offline and print cleanly.
3. **Tabbed SPA, never a scroll.** One focused idea per tab. Lead each tab with its strongest visual.
4. **Dynamise the dynamics.** Any process, lifecycle, or state machine becomes a real SVG flow diagram showing temporal/sequential order, including branches and loop-backs, with clickable nodes that reveal detail. Never explain a flow with side-by-side text columns.
5. **Per-variant diagrams.** Multiple scenarios (user stories, modes) each get their own switchable flow.
6. **Progressive disclosure.** Collapse fine detail (FRs, NFRs, fine print) into accordions; surface only IDs and a one-line gist by default.
7. **Color = signal, not decoration.** Emphasize what matters (gates, decisions, ordered phases); keep routine/native detail gray or omit it. If most of the page is colored, nothing stands out.
8. **Interactivity earns its place.** Every tab, accordion, and click-to-detail must reduce the reviewer's effort. If it does not, remove it.
## Workflow
To build the review experience, follow these steps in order.
### 1. Extract the critical points
Read the source PRD/spec. Identify what a reviewer must understand to approve: the problem and its evidence, the core method/model, the key mechanism (how it works), the distinct usage scenarios, the scope (in/out), and — most important — any **decisions that block sign-off** (look for "DECISION NEEDED", open questions, `[ASSUMPTION]` tags). Note anything that is a *process, lifecycle, or state machine*, since that becomes the centerpiece diagram.
### 2. Choose a tab structure that fits the document
Pick 4–6 tabs. A common, strong mapping is: **The Problem → The Method → The Mechanism → Usage Stories → Build Plan**. Rename to fit the document's vocabulary. The invariant: one focused idea per tab, strongest visual leads, fine print collapses. See `references/diagram-and-color-patterns.md` for the tab-structure rationale.
### 3. Start from the scaffold
Copy `assets/prd-review-scaffold.html` to the output location (next to the source file, e.g. `prd.html` in the same folder). It already contains the tab nav, view-switching JS, SVG arrowhead marker defs, the clickable-node detail-panel pattern, per-variant sub-tabs, accordion CSS, and the color-token system. Replace the `{{PLACEHOLDERS}}` and example views with extracted content. Do not rebuild this boilerplate from scratch.
### 4. Build the mechanism diagram (the centerpiece)
For the "how it works" tab, build an SVG flow/state machine showing the real temporal flow. Lay nodes top-to-bottom in execution order; draw connectors first; branch at gates with a green YES edge continuing down and a **dashed red loop-back** edge returning to an earlier node; pulse the gates; make every node clickable into the detail panel. Consult `references/diagram-and-color-patterns.md` for the exact construction recipe, loop-back path geometry, and marker usage.
### 5. Build per-variant flows
For each usage scenario, add a switchable variant. For mostly-linear scenarios a horizontal pipeline of chips reads fastest; highlight only the node that *differs* between variants and add a one-sentence "What's different" callout mapping to the relevant FR/decision. Reach for full SVG only when a variant has its own branching/loops.
### 6. Apply the color system
Use the scaffold tokens consistently: indigo `--accent` for the actor/governed unit, the `--hot → --warm → --cool` ramp for any ordered sequence (phases/severity/maturity), red `--block` strictly for gates and blocking decisions, green `--go` for pass/done, gray for routine. Default everything to gray and promote to color only where attention is earned. Always include the diagram legend.
### 7. Surface the decisions needing sign-off
On the Build-Plan tab, render the blocking decisions prominently (warn/block callouts) and collapse the full FR/NFR text into accordions showing IDs plus a one-line gist. The reviewer should be able to find "what do I need to decide?" in seconds.
### 8. Open it for review
Open the result in the browser (`open <path>` on macOS) so the user can inspect and approve. State clearly that the `.md` remains the canonical source and the HTML matches it on the critical points only.
### 9. Iterate on feedback
Expect the reviewer to push for clearer flow, less clutter, or stronger emphasis. Treat that as the normal loop: adjust diagrams, demote detail to gray/accordions, sharpen color usage, reopen.
## Bundled resources
- **`assets/prd-review-scaffold.html`** — the reusable single-file HTML scaffold (tab nav, view-switching JS, SVG marker defs, clickable detail panel, per-variant sub-tabs, accordion CSS, color tokens). Copy and fill it; do not recreate it.
- **`references/diagram-and-color-patterns.md`** — detailed techniques: the color-as-signal table, the SVG flow/state-machine construction recipe with loop-back geometry, per-variant flow patterns, tab-structure rationale, and what to leave out. Load when building diagrams or deciding the color system.
Related in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.