What founder video do you want to make? Required:

• Product URL —

  https://...

  (anything with a real homepage)
• Founder — name + role, e.g. "Eli Kim, CEO"
• Founder photo — local path, https URL, OR

  generate

  (I'll create a portrait)

Optional (sensible defaults if omitted): brand kit path · custom on-phone screenshots · music · aspect (16:9 / 9:16 / 1:1) · location image · voice style · product type

Example:

/founder-product-video https://scrapegraphai.com --founder "Eli Kim, CEO" --photo ~/Pictures/eli.jpg

--quick

--config <path>

url

• --config <path>

  points to a JSON file with pre-baked values for the canonical input contract:

  url

  ,

  brand_kit_path

  or

  build_brand

  ,

  founder_name

  ,

  founder_role

  ,

  founder_photo

  ,

  assets

  ,

  music_url

  ,

  aspect_ratio

  ,

  location_image_url

  ,

  voice_style

  ,

  product_type

  , and

  lower_third

  .

• --quick

  means use defaults for optional extras, auto-build the brand kit with

  build-a-brand --quick

  if

  brand-kit

  is omitted, and use

  founder_photo = "generate"

  when no photo is supplied.
• For

  --quick

  or

  --config

  , do not stop for confirmation at the brand-kit branch, founder-photo generation prompt, optional-extras prompt, script choices, or end-card/caption defaults. Record assumptions inline and continue.
• If

  url

  or founder name/role cannot be found in args or config, stop once with a single compact missing-fields list instead of starting a multi-turn Q&A loop.

https://...

• If a path -> use it (

  state.brand_kit_path = <path>

  ). Accept either

  brand.json

  or an exported

  build-a-brand

  kit containing

  brand.md

  ,

  tokens/tokens.json

  , and logo assets.
• If "build" -> invoke the

  build-a-brand

  skill on the URL/brief and wait for the exported brand kit. This is a full identity workflow and may pause for user choices; surface those prompts in interactive mode.
• Fast lane: if config provides

  brand_kit_path

  , use it. If config sets

  build_brand

  or

  --quick

  omits

  brand-kit

  , invoke

  build-a-brand --quick

  on the URL/brief and wait for the exported brand kit; do not surface build-a-brand prompts or stop for identity choices. After either branch, set

  state.brand_kit_path

  . Only stop with a single compact missing-fields list if there is no path and the brand kit cannot be built.

• founder_name

  — e.g. "Semi"

• founder_role

  — e.g. "CEO, Pika"

• founder_photo

  — local path / https URL / OR the literal string

  generate

  to auto-create a portrait. If

  generate

  , prompt the user for a 1-line vibe ("warm, casual smart attire" / "Pixar-style 3D animation" / etc.) — this becomes the seed prompt for

  mcp__pika__generate_image

  in step [4].
• Fast lane: use founder values from args/config. If

  founder_photo

  is omitted, set

  founder_photo = "generate"

  and use a neutral founder-portrait vibe derived from the product tone; do not stop for a separate photo-vibe prompt.

state.lower_third = true

• Custom imagery — list of

  assets

  (product photos / app screenshots) shown on the founder's phone. Default if omitted: use screenshots from

  brand.json.screenshots

  when present, otherwise look for obvious screenshots or product images inside the brand kit. If none exist in interactive mode, ask once whether to proceed without imagery or wait for uploads; in the fast lane, proceed without custom imagery and record the assumption.
• Music — local path / https URL / OR

  generate

  (instrumental, ~60s). Default:

  generate

  via MiniMax.
• Lower-third — optional. Default: off. If enabled, the render is MCP but the overlay onto body video uses one local ffmpeg pass.
• Aspect ratio —

  16:9

  (default),

  9:16

  ,

  1:1

  .
• Location — defaults to a flat seamless backdrop in

  state.brand.colors.accent

  (the brand-accent backdrop pattern validated 2026-05-03 on a Pika.me run where accent happens to be lavender — clean studio-shoot look, character against a single brand color, whatever the brand's accent is). Override with a path / URL / text description if the user wants office, outdoor, etc.
• Voice style — VO direction string for SeeDance, e.g. "warm authentic founder energy, conversational". Default: derived from

  brief.tone

  .
• Product type —

  digital | physical_apparel | physical_object | consumable | service

  . Default: auto-derived in step [2] from asset analyses.

state.inputs

<workdir>/inputs.json

• url

  — the product website (https://). Drives step [1] brief.

• brand_kit_path

  — brand kit folder. Required. End card AND lower-third consume

  brand.json

  when present, otherwise

  brand.md

  ,

  tokens/tokens.json

  , and logo assets from a

  build-a-brand

  export. See step [4.5].

• founder_name

  +

  founder_role

  +

  founder_photo

  — required from intake. Step [4] normalizes

  founder_photo

  into

  founder_photo_url

  and

  character_url

  before any SeeDance call.

• assets

  — optional array of

  { url, role?, caption? }

  . Defaults to the screenshots captured by the brand kit.

  role

  is a hint string mapping the asset to a script beat (

  hero

  ,

  feature_a

  ,

  cta

  , etc.).

• location_image_url

  — optional. Defaults to a generated solid-color backdrop in

  state.brand.colors.accent

  .

• music_url

  — optional. Defaults to

  generate

  (MiniMax instrumental in step [7]).

• aspect_ratio

  — default

  16:9

  .

• voice_style

  — optional, defaults to

  brief.tone

  .

• product_type

  — optional, auto-derived in step [2].

references/ops-notes.md

• Use a unique

  seed

  per SeeDance act (101, 202, 303, 404). Identical generation params can replay cached failures.
• MiniMax music length is controlled by five

  lyrics

  sections (

  intro

  /

  verse

  /

  bridge

  /

  chorus

  /

  outro

  ) with instrumental parentheticals; prompt prose alone is not a reliable length control.
• If SeeDance rejects a real-person founder photo, re-roll the founder ref with stronger stylization rather than retrying the same rejected reference.
• For local brand-kit logos, upload only logo-appropriate raster assets (

  image/png

  ,

  image/jpeg

  , or

  image/webp

  ). Do not send SVGs to

  mcp__pika__upload_asset

  ; choose the PNG export from

  build-a-brand

  or rasterize first.
• Use CSS

  background-image: url(...)

  for CDN-hosted logo/photo assets in end-card HTML;

  <img crossorigin>

  is blocked by CDN CORS.
• Use server-side deterministic tools for captions, concat, and mix. Local ffmpeg is only the fallback for arbitrary transparent lower-third overlay composition, and that fallback is used only when

  state.lower_third = true

  .
• Decompose every 15s act into 3 time-coded sub-shots. Single-shot acts look static.
• Open each act with the style-match location framing and repeat the same

  WARDROBE LOCK:

  sentence across all 4 act prompts.

assets

mcp__pika__analyze_media

analyze_media(
  media: <asset.url>,
  query: 'Describe this product image briefly. Return STRICT JSON:
  {
    "content_description": "1-line summary of what is visible",
    "asset_type": "digital_screen | physical_apparel | physical_object | consumable | infographic | other",
    "key_elements": ["3-5 specific UI elements / features / objects in the image"],
    "visible_copy": "any text visible — heading, button label, tagline, t-shirt graphic text (or empty string)",
    "primary_colors": ["#hex", "#hex", "#hex"],
    "vibe": "1-line visual feeling",
    "best_for_act": "hook | problem | solution | proof"
  }
  Return ONLY the JSON.'
)

asset_type

• digital_screen

  — app UI / website screenshot / SaaS dashboard / mobile app capture

• physical_apparel

  — t-shirts, hoodies, hats, anything wearable (model + garment)

• physical_object

  — gadgets, accessories, packaged goods, anything held in hand

• consumable

  — food, beverages, supplements (something used/eaten/drunk)

• infographic

  — chart, diagram, data viz, illustration

• other

  — anything else; describe and pick best fit

asset_analyses[i]

1. character_voice_profile

   — 3-4 lines describing the character's DEFAULT delivery (carries through every act for consistency)
2. Per-shot

   asset_index

   +

   reveal_beat

   — what asset is visible in this shot and how it's revealed
3. Per-shot

   beats[]

   — line-by-line acting direction with

   emotion

   +

   physical

   + silence beats between sentences

• First-person ownership — "I built", "we shipped", "we use it ourselves", "honestly we just want this everywhere"
• A personal stake or origin moment — Act 1 should reference a frustration the founder lived through, NOT the product abstractly. "Every time I tried building X, I hit the same wall" beats "X is hard."
• Conversational connectives — "look", "honestly", "the thing is", "so", "actually", trailing "..." for thinking. These are throwaway words in writing but the breath of natural speech.
• A "bet" framing for the product — "what if X just worked?", "we asked ourselves", "the whole idea was". Founders frame their product as an answer to a question they asked themselves, not as a list of capabilities.
• One concrete anchor — a specific number, a specific time, a specific scenario. "24 thousand devs starred it last year" beats "it's popular." "At 3 AM the layout breaks" beats "scrapers are unreliable."
• Invitation-energy CTA — "come try us", "go play with it", "we just want it everywhere". NOT "stop scraping. start extracting." (that's a Don Draper tagline, not a founder).

• Act 1: Personal stake / pain. First person. Reference a specific frustration the founder lived through. Land on the problem named cleanly.
• Act 2: The bet. "So we built X. The idea was — what if [pain] just worked?" One sentence on what it actually does (URL → data, prompt → image, etc).
• Act 3: Proof + community. ONE specific number (stars, customers, ARR). One casual mention of integrations or where it's used. Tone: quiet confidence, not bragging.
• Act 4: Invitation. "If [reader's situation] — come try us. [URL]. [One inviting line]." End on warmth, not a tagline.

character_voice_profile

brief.tone

character_image_url

product_type

wardrobe_lock

brief.tone

casual

playful

professional

technical

disruptive / edgy

"Casual confidence, like explaining the connector to a friend at coffee. Slight smirk default. Eyebrow flicks on key reveals. Hand goes to chin when thinking, opens out flat on the big 'meet Pika M C P' claim. Pauses with held eye contact rather than filling silence. Lands punchlines deadpan and lets a small smile arrive a beat after."

"Sharp dry wit. Talks fast in clipped sentences with sudden pauses. Default slight smirk with one raised eyebrow. Eye-rolls on the pain points ('boring', 'generic'). Mischievous grin breaks through on punchlines but disappears immediately. Hands stay mostly still — the FACE does the work."

physical

• A hand or arm gesture (open palm, count on fingers, dismissive flick, point at self/camera, hand to chest, wider arm sweep, hand-to-temple thinking)
• A torso shift (lean forward, lean back, slight body turn, shoulder shift)
• A head action LARGER than a micro-flick (turn left/right and back, tilt 8°+, slow head shake, head bob on rhythm)
• A directional eye flick combined with eyebrow movement (look down then snap up to camera, etc.)

• Silent

  (beat)

  markers between spoken sentences (those are meant to be still — the held look is the point)
• Final landing beat at end of an act when the camera is already moving (camera does the work)

{
  "product_type": "<from step [2]>",
  "character_voice_profile": "Casual confidence, like explaining to a friend at coffee. Slight smirk default. Eyebrow flicks on key reveals. Hand goes to chin when thinking, opens out flat on big claims. Pauses with held eye contact rather than filling silence. Lands punchlines deadpan, lets a small smile arrive a beat after.",
  "segments": [
    {
      "act": 1,
      "shots": [
        {
          "type": "A",
          "asset_index": null,
          "beats": [
            { "text": "Claude is brilliant.", "emotion": "declarative respect — say it like she means it", "physical": "soft direct eye contact, slight nod" },
            { "text": "(beat)", "physical": "subtle smirk arrives, eyes hold camera" },
            { "text": "But it's also kind of...", "emotion": "playful pivot, ellipsis hangs", "physical": "slight head tilt right, eyes drift up briefly on the ellipsis" },
            { "text": "shapeless.", "emotion": "deadpan landing", "physical": "eyes return to camera, single dismissive shrug" }
          ]
        },
        {
          "type": "B",
          "asset_index": null,
          "transition_from_prev": "Camera pushes in slowly from the medium framing into a tighter close-up, drifting slightly off-axis to her right so a different slice of the brick wall and window light is visible behind her",
          "beats": [
            { "text": "No face. No voice. No personality of its own.", "emotion": "staccato dismissal", "physical": "small head shake on each, eyebrow flick on 'personality'" },
            { "text": "(beat)", "physical": "held look, eyes lock camera, soft smile starts to arrive" },
            { "text": "Just an empty assistant waiting for orders.", "emotion": "flat, slightly resigned deadpan", "physical": "neutral face" },
            { "text": "(beat)", "physical": "soft confident smile arrives, lean forward begins" },
            { "text": "That's about to change.", "emotion": "grounded conviction, the turn", "physical": "lock eyes, single confident nod on 'change'" }
          ]
        }
      ]
    },
    {
      "act": 2,
      "shots": [
        {
          "type": "C",
          "asset_index": 0,
          "reveal_beat": "The character holds her phone up toward camera at chest height, screen facing the viewer. Phone screen exactly matches @Image3 — match the screen content exactly, do not invent UI.",
          "beats": [
            { "text": "Meet Pika M C P.", "emotion": "introduction with quiet pride", "physical": "phone lifts to camera, eyes flick from screen to lens" },
            { "text": "One connector that gives your Claude a face, a name, a personality.", "emotion": "warm steady build", "physical": "free hand counts the three on fingers — face, name, personality" }
          ]
        },
        {
          "type": "A",
          "asset_index": null,
          "transition_from_prev": "Camera pulls back from the phone and arcs slightly left, the phone lowers out of frame as we end on a medium shot of the character with the desk and whiteboard now visible behind her",
          "beats": [
            { "text": "And the ability to make videos, images, audio.", "emotion": "expanding the promise", "physical": "open-palm gesture sweeps wider on each item" },
            { "text": "Right inside the chat.", "emotion": "the grounding kicker", "physical": "hand lands flat, eyebrows up, slight smile arrives" }
          ]
        }
      ]
    },
    {
      "act": 3,
      "shots": [
        {
          "type": "D",
          "asset_index": null,
          "beats": [
            { "text": "Setup takes thirty seconds.", "emotion": "matter-of-fact reassurance", "physical": "walks past a desk, glances at a laptop briefly" },
            { "text": "Open Claude settings, paste the URL, sign in.", "emotion": "quick rhythmic checklist", "physical": "counts three on fingers as she walks" }
          ]
        },
        {
          "type": "E",
          "asset_index": 1,
          "reveal_beat": "Close-up of hands holding phone. Screen first matches @Image3 ('What if your Claude could be CAMI'), then transitions mid-shot to @Image4 (setup guide). Match each screen exactly, do not invent UI.",
          "transition_from_prev": "Camera dollies in fast past her shoulder to land on a tight close-up of her hands and the phone, the loft background drops fully out of focus",
          "beats": [
            { "text": "Your Claude becomes Cammy.", "emotion": "the soft surprise reveal", "physical": "small smile at the phone, then up to camera, eyes warm" }
          ]
        },
        {
          "type": "A",
          "asset_index": null,
          "transition_from_prev": "Camera pulls back and tilts up from the phone to find her face in a medium shot, the brick wall and afternoon light now visible behind her on a different side of the loft than Shot D",
          "beats": [
            { "text": "Or whoever you build.", "emotion": "casual aside", "physical": "small shrug, slight smirk" },
            { "text": "(beat)", "physical": "held look, smirk fades into warm sincerity" },
            { "text": "Now talk to her like a person.", "emotion": "the real point — quiet conviction", "physical": "single nod on 'person', eyes hold" }
          ]
        }
      ]
    },
    {
      "act": 4,
      "shots": [
        {
          "type": "F",
          "asset_index": null,
          "beats": [
            { "text": "Skills bundled in.", "emotion": "casual confidence, intro to a list", "physical": "slight tilt of the chin, knowing look" },
            { "text": "Podcasts. Explainer videos. U G C ads.", "emotion": "rhythmic three-beat list", "physical": "small nod on each, eyebrow flick on 'U G C'" },
            { "text": "All from chat.", "emotion": "the grounding tag", "physical": "open-palm gesture lands flat, slight smile arrives" }
          ]
        },
        {
          "type": "B",
          "asset_index": null,
          "transition_from_prev": "Camera pushes in slowly from the wider brand-context framing into an intimate medium close-up; the environment recedes into soft bokeh, the character fills more of the frame",
          "beats": [
            { "text": "So stop wrestling with generic A I.", "emotion": "direct address, low-key challenge", "physical": "raised eyebrow, slight head tilt" },
            { "text": "(beat)", "physical": "held look, smirk grows" },
            { "text": "Pikafy your Claude.", "emotion": "the brand line, said with certainty", "physical": "lean slightly into camera, lock eyes" },
            { "text": "Start at pika dot me slash M C P.", "emotion": "warm CTA, the invitation", "physical": "soft confident smile, single closing nod on 'M C P'" }
          ]
        }
      ]
    }
  ]
}

1. For digital products: only shots

   C

   and

   E

   get an

   asset_index

   (phone-reveal moments). Other shots show the founder without a specific UI reference.
2. For physical products: any shot where the product is visible in frame gets an

   asset_index

   . The asset is the source of truth for what that product looks like. Acts can reuse the same asset across multiple shots, OR show a different asset per shot to demo product variety.
3. For service products: no

   asset_index

   anywhere — the script relies on dialogue + environment.

@Image3

@Image4

asset_analyses[i].key_elements

asset_analyses[i].visible_copy

product_type

• ❌ "Two-shot" → adds a SECOND PERSON to the frame (term means "shot with two subjects" in film, but SeeDance just sees "two" + "person").
• ❌ "Three-shot" — same trap.
• ❌ "Over-shoulder" / "OTS" → adds a phantom shoulder/back-of-head in the foreground for the character to interact with. The character then performs toward that phantom person, not toward the camera. (Verified 2026-05-02 — caused Semi to "show her phone to someone in OTS" in an early Pika MCP test.)
• ❌ "Master shot" — can be misread as "the master / their boss".
• ✅ "Medium shot", "Close-up", "Wide" are safe; they're commonly used colloquially.

• "Over-shoulder reveal" → "the character holds her phone up toward camera, screen facing the viewer at chest height"
• "Two-shot" → "wide framing of the character with [product/logo/environment] in frame"
• "POV" → "low camera angle from the character's eyeline"

physical_apparel

physical_apparel

digital

physical_apparel

physical_object

consumable

digital

physical_apparel

physical_object

consumable

digital

physical_apparel

physical_object

character_url

• If

  founder_photo

  is an HTTPS URL, set

  founder_photo_url = character_url = founder_photo

  .
• If

  founder_photo

  is a local path, upload it with

  mcp__pika__upload_asset

  , then set

  founder_photo_url = character_url = public_url

  .
• If

  founder_photo

  is

  generate

  , call

  mcp__pika__generate_image

  and use the returned URL:

generate_image(
  prompt: "<founder vibe>. Professional founder portrait, clean studio lighting, sharp focus on face, confident expression, suitable as character reference for video generation.",
  aspect_ratio: "3:4",
  resolution: "2K"
)

• If

  location_image_url

  is an HTTPS URL, set

  location_url = location_image_url

  .
• If it is a local path, upload it with

  mcp__pika__upload_asset

  and set

  location_url = public_url

  .
• If it is a text description, generate a custom location reference with

  mcp__pika__generate_image

  .
• If no custom location was supplied, do nothing here. Step [4.5] generates the default brand-accent backdrop after

  state.brand

  exists.

state

brand_kit_path

build-a-brand

build-a-brand --quick

brand.json

build-a-brand

• brand.md

  for name, tagline, voice, typography names, and logo descriptions.

• tokens/tokens.json

  for colors and font tokens.

• logo/

  assets for wordmark, symbol/icon, and lockups.

state.brand

state.brand

name

brand.json.name

brand.md

wordmark_path

logo.wordmark.path

logo/wordmark/*.{png,jpg,jpeg,webp}

mcp__pika__upload_asset

public_url

state.brand.wordmark_url

icon_url

logo.icon_mark.path

logo/symbol/*.{png,jpg,jpeg,webp}

mcp__pika__upload_asset

public_url

state.brand.icon_url

colors.primary

ink_primary

surface_dark

tokens.color.text

colors.surface

surface_page_bg

surface_white

tokens.color.background

colors.accent

tokens.color.primary

colors.highlight

fonts.display_family

brand.md

fonts.text_family

brand.md

fonts.mono_family

mcp__pika__render_html_animation

mcp__pika__upload_asset

public_url

state.brand

mcp__pika__upload_asset

image/svg+xml

mcp__pika__html_to_png

<svg>

public_url

location_url

mcp__pika__html_to_png

state.brand.colors.accent

aspect_ratio

16:9

9:16

1:1

file_url

location_url

html_to_png(
  html: "<body style='margin:0;background:{accent}'></body>",
  format: "png",
  mode: "sync",
  raster_options: { viewport_px: { width: W, height: H }, device_scale: 1 }
)

@Image1

@Image1

@Image2

1. Character + location identity (always the same opening line).
2. Character voice — the act prompt repeats

   script.character_voice_profile

   verbatim. This carries the actor's personality through every shot.
3. Per-shot blocks — each shot gets its composition (or

   reveal_beat

   if defined), then a per-line "Acting beats" list built from the shot's

   beats[]

   , then the dialogue inside

   <<<voice_1>>>

   .

WARDROBE LOCK:

Background context:

<<<voice_1>>>

Transition:

Hard cut:

Native lip-synced dialogue audio, no music overlay.

## Load-bearing phrases

The character (matching @Image1) in a setting whose visual style, palette, lighting and materials match @Image2.

CHARACTER VOICE: {script.character_voice_profile}
WARDROBE LOCK (verbatim across all 4 acts): {script.wardrobe_lock}

Shot {first}: {if shot.reveal_beat exists: shot.reveal_beat ELSE: composition + camera from Shot table}. Background context: {distinct physical position in the space — which wall / window / feature is behind the character}.
Acting beats:
  • "{beat[0].text}" — {beat[0].emotion}; {beat[0].physical}.
  • (silence) — {silence beat physical}.
  • "{beat[N].text}" — {beat[N].emotion}; {beat[N].physical}.
<<<voice_1>>>{joined beat texts excluding (beat) markers, with periods and commas as written}<<<voice_1>>>

[If 2+ shots in this act:] Transition: {shot[1].transition_from_prev}.
Shot {second}: {composition or reveal_beat}. Background context: {a DIFFERENT physical position from Shot {first} — different wall / different angle / different background feature}.
Acting beats: ...
<<<voice_1>>>...<<<voice_1>>>

[If 3 shots:] Transition: {shot[2].transition_from_prev}.
Shot {third}: ... Background context: {a THIRD distinct physical position — must be visually different from Shots {first} and {second}}.

Native lip-synced dialogue audio, no music overlay.

• Open with "The character (matching @Image1) in a setting whose visual style ... match @Image2" — never "inside the location matching @Image2". SeeDance reads the LITERAL backdrop from @Image2 if you say "inside the location" — see Known infra quirks.
• Don't describe the character in prose — SeeDance reads visual identity from

  @Image1

  . EXCEPTION: lock wardrobe explicitly via the

  WARDROBE LOCK

  line (e.g. "wearing the same charcoal hoodie over a dark band tee throughout"). @Image1 alone doesn't lock wardrobe across separate 15s generations.
• Don't add aesthetic adjectives that contradict the references. If

  @Image1

  is a 3D Pixar-style character, don't add "photorealistic" anywhere in the prompt.
• The CHARACTER VOICE + WARDROBE LOCK lines appear once per prompt, before the shots. They prime SeeDance for the actor's whole vibe + clothing continuity.
• Each shot block includes a

  Background context:

  line describing a distinct physical position in the space — different wall, different window, different background feature than the other shots in this act and ideally distinct from the other acts. This forces SeeDance to render scene variety while keeping the brand aesthetic anchored to @Image2.

• (beat)

  markers stay OUT of the

  <<<voice_1>>>

  payload. They're acting direction only — the natural pause between sentences in the dialogue text is where they happen.
• Every shot beyond the first in an act gets a

  Transition: …

  line built from

  shot.transition_from_prev

  . This narrates the camera move between framings, forcing SeeDance to render real spatial motion (different parts of the room behind the character) instead of an unmotivated re-frame that reads as a jump zoom.

(0-4s)

(4-9s)

(9-15s)

1. A different camera framing — never two consecutive sub-shots with the same shot type. Mix medium / close-up / wide / lower-angle. The user reads variety as production value.
2. A different camera motion — push-in, pull-back, lateral arc, orbit, handheld, static-held, rack-focus. Not all "slow push-in".
3. A different physical position in the space — see Location-reference rule in "Known infra quirks". Each shot must describe a different wall / window / feature visible behind the character so SeeDance moves through the space instead of reproducing one literal backdrop.
4. An ENERGETIC physical action per beat — see [3c.1] above. Hand gestures, leans, head turns, shoulder shifts. No facial-only beats.
5. The dialogue sub-portion for that window, wrapped in

   <<<voice_1>>>...<<<voice_1>>>

   tokens inside the shot's block.

generate_reference_video(
  provider: "seedance",
  resolution: "1080p",
  aspect_ratio: "16:9",
  duration: 15,
  seed: 101,                   # unique per act (101, 202, 303, 404) — busts the idempotency cache
  sound: true,                 # native lip-sync from each <<<voice_1>>> block
  reference_images: [character_url, location_url, ...assets_for_this_act],
  prompt: <full prompt — see template in section above, 3 time-coded sub-shots inline>
)

• Retry the missing act once with the same prompt,

  reference_images

  ,

  duration

  ,

  sound

  ,

  resolution

  , and

  aspect_ratio

  , but a new seed (

  original_seed + 1000

  ). Do not rerun successful acts.
• If the retry completes, insert that URL into the original act slot and continue with

  act_urls = [act1, act2, act3, act4]

  .
• If the retry cannot complete, stop and surface the upstream SeeDance timeout. You may return completed act URLs as a diagnostic preview, but do not deliver a partial concat as

  final_url

  , do not call it production-ready, and do not proceed to step [6].

• provider: "minimax-music"

• lyrics

  field: exactly 5 sections (

  [intro]

  /

  [verse]

  /

  [bridge]

  /

  [chorus]

  /

  [outro]

  ), separated by

  \n\n

• Each section contains a

  (instrumental — …)

  parenthetical
• No real English words inside any section (MiniMax sings them)
• No

  length

  /

  duration

  extras (the wrapper drops them; section count is the actual length lever)

generate_music(
  provider: "minimax-music",
  prompt: "<one-line overall description: register + headline instrument + mood + BPM>",
  lyrics: "[intro]\n(instrumental — <opening texture: 2-3 instruments, sparse>)\n\n[verse]\n(instrumental — <add a layer: percussion or bass or counter-melody>)\n\n[bridge]\n(instrumental — <peak tension: layer swell or chord shift>)\n\n[chorus]\n(instrumental — <full arrangement, the emotional payoff>)\n\n[outro]\n(instrumental — <gentle resolve, instruments fall away>)"
)

• [section]

  tags scale duration: 5 sections → ~60–80s output. Use 5 for a 60s body + 5s end card.

• (instrumental — …)

  parenthetical inside each section is read as a production directive ("no vocals, play this arrangement"). Without it, MiniMax may sing.

• prompt

  field carries the OVERALL style snapshot. Section parentheticals do the per-beat work.

music_url

result.duration_seconds

• If

  >= 50s

  → mix it. Expected path with the 5-section structure.
• If

  < 50s

  → re-roll with the same call. After 2 attempts, accept whatever returned —

  mcp__pika__edit_audio_mix

  plays the music for its duration then leaves silence; dialogue carries the rest.

• ❌ Bare

  lyrics: "[instrumental]"

  — MiniMax sings the literal word "instrumental" for ~10s.
• ❌ Omitting

  lyrics

  entirely — output drops to ~17–30s.
• ❌ Putting the duration in the

  prompt

  text ("90 second instrumental") — has no effect; section count drives length.
• ❌ Real lyrics with English words inside any section — MiniMax sings them.
• ❌ Copying the same piano-and-pads example for every brand — the recipe is the pattern, not the sound. Pick a register per brand.

Pipeline ordering note — music mix happens in step [10], AFTER end-card concat. Mixing music into the body before the end card is concatenated leaves the end card silent (the music track ends at the cut). Always: overlays on body → end card → concat → THEN mix music over the full assembled clip.

mcp__pika__add_captions

mcp__pika__render_html_animation

mcp__pika__edit_pip