Skip to content

majiayu000/spellbook / skills/build-product-demo

build-product-demo

Verified

Plan, produce, or diagnose evidence-backed product demo videos and screen-recorded promotional walkthroughs. Use when the user asks to make a product demo, launch video, feature showcase, app walkthrough, demo reel, or polished recording; wants every important capability shown without a slow feature tour; asks to remove dead time or fix narration-to-action pacing; or needs a repeatable script, capture plan, and verified final media. Do not use for fictional commercials with no real product proof, general-purpose video editing, or documentation-only tutorials that do not need a promotional narrative.

FileKindSize
SKILL.mdskill md13.0 KB
agents/openai.yamlconfig239 B
evals/evals.jsonconfig5.0 KB
references/artifact-contract.mdmarkdown6.8 KB
references/directing.mdmarkdown6.6 KB
references/production.mdmarkdown6.0 KB
scripts/analyze_demo_pacing.pyexecutablescript15.5 KB
scripts/probe_demo_media.pyexecutablescript6.8 KB
scripts/validate_demo_plan.pyexecutablescript16.3 KB

references/artifact-contract.md

View on GitHub
# Product Demo Artifact Contract

Use a dedicated demo directory outside generated build output. Preserve raw
captures and validation evidence so the result can be reviewed or recut.

## Recommended package

```text
<demo-dir>/
├── brief.md
├── feature-coverage.md
├── SCRIPT.md
├── beat-plan.json
├── evidence/
├── recording/
├── final/
│   └── <product>-demo.mp4
└── verification.json
```

Planning mode requires the four top-level planning files. Production mode also
requires raw evidence or pointers, the final artifact, and verification output.
Do not create empty placeholder directories.

## `brief.md`

Record:

- public or sanitized product repository identity, exact revision, runtime, and
  date; keep the operational checkout path outside the production package;
- audience, distribution channel, duration, aspect ratio, and language;
- audience doubt and proof proposition;
- opening, turning, and landing state;
- truth boundary and production simplification rule;
- assumptions, blockers, exclusions, credentials, paid-provider, publishing,
  and destructive-action boundaries.

## `feature-coverage.md`

Use one row per candidate capability:

| Capability | Audience value | Evidence level | Demo role | Visible proof | Included | Reason |
|---|---|---|---|---|---|---|

Do not treat documentation copy as live evidence.

## `SCRIPT.md`

Include the director treatment, chapter summaries, one narration block per
beat, on-screen text, sound intent, and disclosure language. Keep narration and
visible action adjacent for review.

## `beat-plan.json`

Use this shape:

```json
{
  "schema_version": 2,
  "product": {
    "name": "Example",
    "repository": "public URL or sanitized repository identity",
    "revision": "exact revision"
  },
  "audience": "specific audience",
  "proof_proposition": "observable claim the demo will prove",
  "reference_benchmark": "path or identity of the comparison artifact",
  "duration_seconds": 5,
  "max_information_gap_seconds": 5,
  "max_attention_gap_seconds": 3,
  "first_product_action_seconds": 4,
  "native_surface_target_ratio": 0.7,
  "delivery": {
    "width": 1440,
    "height": 900,
    "fps": 30,
    "container": "mp4",
    "video_codec": "h264",
    "audio_codec": "aac",
    "duration_tolerance_seconds": 0.25
  },
  "truth_boundary": {
    "live": ["real product paths shown"],
    "deterministic": ["fixtures or providers used only for stability"],
    "composite": ["editorial combinations"],
    "excluded": ["claims not made"]
  },
  "beats": [
    {
      "id": "result-hook",
      "type": "normal",
      "start_seconds": 0,
      "end_seconds": 5,
      "claim": "The product produces a real result.",
      "visible_action": "Show the finished result, then reveal its source.",
      "audience_before": "The audience does not know whether the product works.",
      "audience_after": "The audience has seen the concrete outcome.",
      "entry_state": "Result hidden.",
      "exit_state": "Result visible and attributable.",
      "narration": "Start with the outcome.",
      "audio_intent": "Narration resolves as the result lands.",
      "truth_mode": "live",
      "surface": "native",
      "events": [
        {"at_seconds": 0.8, "kind": "product_action", "description": "Command executes"},
        {"at_seconds": 2.2, "kind": "result", "description": "Native result appears"}
      ],
      "evidence": ["evidence/result.json"],
      "cut_reason": "Return to the starting state to explain the cause."
    }
  ]
}
```

Required top-level fields:

- `schema_version`: integer `2`;
- `product.name`, `product.repository`, `product.revision`: non-empty strings;
- `audience`, `proof_proposition`, `reference_benchmark`: non-empty strings;
- `duration_seconds`, `max_information_gap_seconds`,
  `max_attention_gap_seconds`: positive numbers;
- `first_product_action_seconds`: number from zero through five;
- `native_surface_target_ratio`: number from `0.6` through `1.0`; a lower target
  is allowed only when `native_surface_exception` is a non-empty explanation of
  why the medium makes the ratio inapplicable;
- `native_surface_exception`: omit it normally; use a non-empty string only for
  the documented lower-target exception;
- `delivery.width`, `delivery.height`: positive integers;
- `delivery.fps`: positive number;
- `delivery.container`, `delivery.video_codec`, `delivery.audio_codec`: non-empty
  strings matching the target channel's delivery requirements;
- `delivery.duration_tolerance_seconds`: positive number, normally `0.25`;
- `delivery.duration_tolerance_reason`: required non-empty string when the
  tolerance exceeds `0.25`;
- `truth_boundary.live`, `deterministic`, `composite`, `excluded`: arrays;
- `beats`: non-empty ordered array.

Required beat fields:

- `id`: unique non-empty string;
- `type`: `normal`, `title`, or `hold`;
- `start_seconds`, `end_seconds`: contiguous, non-overlapping times;
- `claim`, `visible_action`, `audience_before`, `audience_after`,
  `entry_state`, `exit_state`, `audio_intent`, `cut_reason`: non-empty strings;
- `narration`: string; it may be empty when sound or silence carries the beat;
- `truth_mode`: `live`, `deterministic`, `composite`, or `title`;
- `surface`: `native`, `composite`, or `title`;
- `events`: ordered array of observable events. Normal beats require at least
  one event. Each event has an absolute `at_seconds`, `kind`, and description;
  kinds are `product_action`, `input`, `result`, `reveal`, `cut`, `sound`, or
  `hold`;
- `evidence`: array of non-empty strings; it may be empty only for `title`;
- `hold_reason`: required for `hold` beats.

The first beat starts at zero. Beats cover the declared duration without gaps
or overlaps. Every beat changes audience knowledge or product state. A normal
beat may not exceed `max_information_gap_seconds`; split it at the next real
action, reveal, or consequence instead of adding decorative cuts.

The native-surface duration must meet `native_surface_target_ratio`, and title
plus composite surfaces must stay below 20% unless
`native_surface_exception` records why the medium makes those ratios
inapplicable. The first
`product_action` event on a normal native-surface beat must occur by
`first_product_action_seconds`. When a documented lower-target
`native_surface_exception` is active, a `product_action` event on a normal
composite proof surface also qualifies. Title events never satisfy this gate.
Consecutive non-hold events may not exceed
`max_attention_gap_seconds`.

## `verification.json`

Record the plan validation result, media probe output, pacing analyzer output,
native/title/composite duration ratios, dense-frame and continuous-playback
review, exact motivated-hold intervals and their analyzer exemptions, retained
evidence paths, known limitations, and final verdict. Do not
mark the demo complete when any required check is missing.
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall