Prerequisites

Install and load these skills before generating (skip if already in context via @pruna):

Skill	Description	Install
`generation-diversity`	Use when writing any generative prompt — ritual seed, explicit structure, scenario axes, and quality gates before paid API calls.	`npx skills add PrunaAI/pruna-skills@generation-diversity -y`
`video-prompting`	Use when crafting video or motion prompts for any generative model — dramaturgy, camera, physics-safe motion, frame anchors, and clip chaining.	`npx skills add PrunaAI/pruna-skills@video-prompting -y`
`audio-prompting`	Use when crafting TTS, music, or bed prompts for any generative audio model — director style, song structure, and post-production layering.	`npx skills add PrunaAI/pruna-skills@audio-prompting -y`
`pruna-api`	Use before any Pruna or Replicate HTTP call — credentials, upload/poll/download, parallel batches, and agent safety.	`npx skills add PrunaAI/pruna-skills@pruna-api -y`

Or install the full suite once: npx skills add PrunaAI/pruna-skills@pruna -y

Follow each skill's Before generating / craft sections — do not restate guide content here.

Agent habit

Routing: `p-video-2-pro` is the cinematic generation clip (text / first frame / first+last frame) with generated audio. Use `p-video-2` when the brief needs 1080p, imported audio, or draft previews. Use `p-video` for cheaper, faster drafts. Talking-head-only (no native scene audio) → `p-video-avatar`. Instruction edit of existing footage → `p-video-edit`.

In the first reply, name `p-video-2-pro` in backticks, confirm PRUNA_API_KEY (or stop with signup links from pruna-api), then ask for required inputs. Open intake → generation-diversity clarification intake before the first POST. When drafting motion prompts, follow Prompt craft (dynamic + faithful) and video-prompting (p-video-2-pro-prompting) — do not paste skill examples. Redirect when When NOT to use fits better.

Prompt craft (dynamic + faithful)

Every input.prompt must be fresh and specific, and must match the user's beat. Diversity never overrides the brief. Craft deltas vs p-video-2: video-prompting (p-video-2-pro-prompting). Write sound or dialogue into the prompt — this model has no audio upload.

Do	Don't
Run the `generation-diversity` random seed ritual; state it; rotate ≥2 free axes (camera move, lighting shift, texture, pacing) when the brief allows	Copy curl examples from this skill (`sports car`, `person turns`, …) or reuse a prior session's prompt
Lock user-required facts first (subject, action, scene; OPEN/MID/CLOSE when frame-anchored; spoken line / SFX when audio is part of the brief)	Swap the subject or motion for a “cooler” clip that ignores the request
Structure with `video-prompting` dramaturgy — subject + action + scene; add camera, lighting, style, and audio intent for locked-in finals	Vague mood-only strings (`cinematic vibe, neon energy`) or a silent prompt when the brief asked for speech or score
Keep speakers to one or two; name the line in the prompt	Crowd a clip with 3+ talking faces, or send `audio` / `draft` / `fps` (not on this API)
Show the drafted prompt + `mode` / `prompt_upsampler` / `duration` / `resolution` before `POST` when the user has not locked wording	Silent regen with a different subject or beat than approved

Fidelity check (before pay): if you remove the user's named subject/action/setting from the prompt, the job is wrong — rewrite. Free axes only fill what the brief left open.

When showing a drafted prompt, still name `p-video-2-pro` (guides help craft; this tool owns the call).

Skill boundary

This skill = one p-video-2-pro prediction per invocation.

Out of scope (do not execute from this skill):

Multi-scene assembly, concat, subagent orchestration, or parallel scene batches
Imported audio / mixed VO / 1080p / draft previews → p-video-2
Motion transfer from a template video → p-video-animate
Talking-head-only / lip-sync host without native scene audio → p-video-avatar
Instruction edit of existing footage → p-video-edit

If the request exceeds one clip, stop and recommend: image-to-video (one beat), visual-transition-reel (multi-scene visual), or narrated-multi-scene (multi-scene + uploaded VO — that path uses p-video-2).

When NOT to use

Use a different skill instead:

Skill	Description	Install
`p-video-2`	Use when someone wants a polished short clip from text, images, or imported audio — 1080p B-roll, start/end frame animation, or a motion shot with a mixed track. Not for cinematic generated-audio clips or talking-head-only hosts.	`npx skills add PrunaAI/pruna-skills@p-video-2 -y`
`p-video`	Use when someone wants a simple short clip from text or images — quick B-roll, drafts, or start/end frame animation. Not when the brief needs cinematic generation, highest quality, tight lip-sync, or imported audio at 1080p.	`npx skills add PrunaAI/pruna-skills@p-video -y`
`p-video-avatar`	Use when someone wants a person on camera speaking a script — lip-synced host, spokesperson, or narrated avatar from a portrait photo.	`npx skills add PrunaAI/pruna-skills@p-video-avatar -y`
`p-video-animate`	Use when someone wants a photo to move like another video — motion transfer, dance remixes, or performance variations from a template clip.	`npx skills add PrunaAI/pruna-skills@p-video-animate -y`
`p-video-replace`	Use when someone wants to swap a person, outfit, or product inside existing footage while keeping the camera move and audio.	`npx skills add PrunaAI/pruna-skills@p-video-replace -y`
`p-video-edit`	Use when someone wants to edit an existing video with a text instruction — recolor, restyle, remove or add objects, change environment or lighting, update on-screen text, or apply optional reference-guided product and accessory edits. Not for a new clip from scratch or ffmpeg assembly.	`npx skills add PrunaAI/pruna-skills@p-video-edit -y`
`video-editing`	Use when assembling or polishing already-rendered clips with ffmpeg — concat, crossfades, burned captions and subtitles, text/logo overlays, before/after sliders, background music beds, platform export — or when composing a multi-layer HTML combination video with Hyperframes. Not for AI video generation, prompt craft, or model-based video edits.	`npx skills add PrunaAI/pruna-skills@video-editing -y`

HTTP (curl)

Upload for image-to-video / frame anchors

curl -X POST "https://api.pruna.ai/v1/files" \
  -H "apikey: ${PRUNA_API_KEY}" \
  -F "content=@/path/to/first-frame.png"

Pass urls.get as input.image (first frame) and/or input.last_frame_image (last frame). Do not upload audio for this model.

Create (async text-to-video — recommended)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -d '{
    "input": {
      "prompt": "A sports car drifting through a neon-lit city at night, cinematic aerial shot",
      "duration": 5,
      "resolution": "768p",
      "aspect_ratio": "16:9"
    }
  }'

Poll and download: follow pruna-api.

Complete the random seed ritual from generation-diversity before writing prompts — do not pass the ritual string as API seed.

Billing is per returned second (not the requested duration). Rate limit: 250 requests per minute.

Resolution	`mode=cost`	`mode=speed` (default)	`mode=quality`
480p	$0.01 / s	$0.02 / s	$0.04 / s
768p	$0.025 / s	$0.035 / s	$0.075 / s

cost is the same output quality as speed, but cheaper and slower. prompt_upsampler does not change the price.

Image-to-video (first frame)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -d '{
    "input": {
      "prompt": "The camera slowly pushes in, the person turns their head and smiles",
      "image": "https://api.pruna.ai/v1/files/FIRST_ID",
      "duration": 5,
      "resolution": "768p"
    }
  }'

First / last frame (visual transition)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -d '{
    "input": {
      "prompt": "Start on the still product hero. The camera holds, then a slow push-in as light moves across the metal. End exactly on the last-frame packshot, label readable, no extra props.",
      "image": "https://api.pruna.ai/v1/files/START_ID",
      "last_frame_image": "https://api.pruna.ai/v1/files/END_ID",
      "duration": 8,
      "resolution": "768p",
      "mode": "speed",
      "prompt_upsampler": "turbo"
    }
  }'

There is no scene-anchor triple on this model (audio is not exposed). Uploaded VO / music → p-video-2.

Create (sync — quick test only)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -H 'Try-Sync: true' \
  -d '{
    "input": {
      "prompt": "A sports car drifting through a neon-lit city at night, cinematic aerial shot",
      "duration": 5,
      "resolution": "480p",
      "mode": "cost"
    }
  }'

Before generating

Complete Prerequisites guide reading order (generation-diversity → video-prompting).
Ritual seed → draft a dynamic + faithful motion prompt (section above) → confirm mode (T2V / I2V / frame pair), duration (5–15s, default 5), resolution (480p / 768p), mode (cost / speed / quality), prompt_upsampler, and prompt with the user.
Pruna notes: when image or last_frame_image is set, aspect_ratio is ignored (canvas follows the stills). Output is 24 fps — do not send fps. Output includes generated audio; write sound or dialogue in the prompt. Do not send audio, draft, save_audio, or prompt_upsampling. Iterate in mode: cost when price is the priority (same quality as speed, but cheaper and slower) or mode: speed when wall-clock time matters (default); then final with mode: quality when fidelity matters. If the request is multi-scene — stop (see Skill boundary). Native 4K is not supported. 1080p / imported audio / draft → p-video-2.

Required input

prompt (string)

Common optional fields

Field	Role
`prompt_upsampler`	Expand the prompt before generation: `off`, `turbo` (default), or `max`. Independent of `mode`
`image`	First-frame reference; when set, `aspect_ratio` is ignored (`jpg`, `jpeg`, `png`, `webp`)
`last_frame_image`	Optional end-frame still
`duration`	5–15s (default 5)
`resolution`	`768p` (default) or `480p`
`mode`	Generation recipe: `cost` (same quality as `speed`, but cheaper and slower), `speed` (default, faster), or `quality` (slower, higher-fidelity)
`aspect_ratio`	When no reference image: `16:9` (default), `9:16`, `4:3`, `3:4`, `3:2`, `2:3`, `1:1`
`seed`	Integer for a reproducible rerun; omit for random

Do not send: audio, draft, fps, save_audio, prompt_upsampling.

Related skills:

Skill	Description	Install
`p-video-2`	Use when someone wants a polished short clip from text, images, or imported audio — 1080p B-roll, start/end frame animation, or a motion shot with a mixed track. Not for cinematic generated-audio clips or talking-head-only hosts.	`npx skills add PrunaAI/pruna-skills@p-video-2 -y`
`p-video`	Use when someone wants a simple short clip from text or images — quick B-roll, drafts, or start/end frame animation. Not when the brief needs cinematic generation, highest quality, tight lip-sync, or imported audio at 1080p.	`npx skills add PrunaAI/pruna-skills@p-video -y`
`gemini-3.1-flash-tts`	Use when someone needs spoken narration or voiceover — explainer tracks, documentary lines, or voice to pair with generated video.	`npx skills add PrunaAI/[email protected] -y`
`image-to-video`	Use when someone wants one short film beat from images — a narrated scene, story moment, or cinematic B-roll with optional voiceover.	`npx skills add PrunaAI/pruna-skills@image-to-video -y`
`visual-transition-reel`	Use when someone wants a montage with transitions between shots — action-sequence reel or multi-scene piece where narration is optional.	`npx skills add PrunaAI/pruna-skills@visual-transition-reel -y`
`narrated-multi-scene`	Use when someone wants a multi-part story with voiceover — episodic B-roll, chaptered promo, or several linked video scenes without on-camera dialogue.	`npx skills add PrunaAI/pruna-skills@narrated-multi-scene -y`
`p-video-avatar`	Use when someone wants a person on camera speaking a script — lip-synced host, spokesperson, or narrated avatar from a portrait photo.	`npx skills add PrunaAI/pruna-skills@p-video-avatar -y`
`p-video-animate`	Use when someone wants a photo to move like another video — motion transfer, dance remixes, or performance variations from a template clip.	`npx skills add PrunaAI/pruna-skills@p-video-animate -y`
`video-editing`	Use when assembling or polishing already-rendered clips with ffmpeg — concat, crossfades, burned captions and subtitles, text/logo overlays, before/after sliders, background music beds, platform export — or when composing a multi-layer HTML combination video with Hyperframes. Not for AI video generation, prompt craft, or model-based video edits.	`npx skills add PrunaAI/pruna-skills@video-editing -y`

PrunaAI/pruna-skills

pruna-skills 是什么？

在你喜欢的 AI 中提问

文档