/video-gen

name: video-gen
description: |
  AI video generation via Fal.ai, Seedance 2.0, Kling, MiniMax Hailuo, xAI Grok Imagine, Google Gemini Omni, OFox, and Agnes Video. Use when the user wants to generate a video clip — text-to-video, image-to-video, first/last-frame transitions, reference-guided generation, multi-shot, or generatively editing / extending an existing clip.
user-invocable: true

Video Gen

Submits one video generation job per call and returns a `jobId`. Job management (wait / status) belongs to `track_progress`; this skill does **not** place videos on the timeline automatically.

When to Use

Any time the user wants to generate a video clip — text-to-video, image-to-video, first-last-frame transition, reference-based generation, multi-shot storyboard, or generatively editing / extending an existing video (producing new generated footage based on a source clip; not timeline trimming).

Models

| Model | Reference | Strengths | | --- | --- | --- | | `fal` + `falModel` | [references/fal.md](references/fal.md) | Explicit Fal catalog; see tool schema for per-model limits | | `seedance2` | [references/seedance2.md](references/seedance2.md) | Default when configured. Multimodal refs, first/last, edit/extend/bridge, 2–15s, 480p/720p/1080p/4k, audio/seed/camera/watermark/last-frame/task controls. | | `kling` | [references/kling.md](references/kling.md) | Technical camera/performance; Omni multi-shot; images ≤7 (≤4 with one feature `refVideos`); std/pro; 3–15s. | | `hailuo` | [references/hailuo.md](references/hailuo.md) | MiniMax 海螺. T2V / I2V / first+last; **6s or 10s**; 512P (Hailuo-02), 720p→768P, 1080P (6s); no multi-ref / multi-shot. | | `grok-imagine-video` | [references/grok-imagine-video.md](references/grok-imagine-video.md) | xAI Grok Imagine. Text-to-video only; 1–15s; 480p/720p/1080p; audio track included. | | `gemini-omni` | [references/gemini-omni.md](references/gemini-omni.md) | Google Gemini Omni Flash (`GEMINI_API_KEY`). T2V / I2V / first+last / ≤6 refImages; 16:9\|9:16; 360p–4k; 3–15 s (first+last 3–10). | | `ofox` | [references/ofox.md](references/ofox.md) | OFox multi-model gateway (Seedance/Wan and more behind one key). Text/image-to-video, first+last frame, up to 9 image refs; 2–30s with per-model API limits; 480p/720p/1080p per model. | | `agnes-video` | [references/agnes-video.md](references/agnes-video.md) | Agnes Video 2.5 Flash. Text, first and/or last frame, or up to 5 images plus 3 audios. 4–12s, 720p only. Frames and references cannot be combined. |

**IMPORTANT:** Before generating, READ the chosen model's reference for capabilities, input channels, modes, prompt structure, and model-specific behavior. Never invent params the reference forbids.

Model Selection

Respect **configured vendors** from the capabilities prompt (only call a model whose key is on).

1. **User named a vendor** ("用海螺", "MiniMax", "Kling", "Seedance") → that `model`, if configured. 2. If Fal.ai is selected or requested, use `model: "fal"` and the requested `falModel` or saved Fal default from capabilities. Ask if no Fal model is selected. Read the Fal catalog constraints in the tool schema; the native-provider limits below do not apply. 3. Else **default `seedance2`** when Seedance is configured. 4. Else if only Kling is on → `kling`. Else if only MiniMax is on → `hailuo`. Else if only xAI is on → `grok-imagine-video`. Else if only Gemini Omni is on → `gemini-omni`. Else if only OFox is on → `ofox`. Else if only Agnes is on → `agnes-video`. 5. Switch away from default when:

• Need **multi-shot customize / intelligence** → `kling` (confirm if not user-named).
• Need **rich multi-modal refs** (video/audio refs, edit/extend) → `seedance2`.
• Need a **short single beat** and only MiniMax is available, or user wants Hailuo → `hailuo` with duration 6 or 10.
• User wants Google / Omni / AI Studio video → `gemini-omni` when configured.

If the required model is **not configured**, say so and offer: another configured video vendor, upload, or Motion Graphic — do not pretend the API exists.

Briefly tell the user what you will generate before submitting.

Tool Params

| Param | Values | Default | | --- | --- | --- | | `model` | `seedance2`, `kling`, `hailuo`, `grok-imagine-video`, `gemini-omni`, `ofox`, `agnes-video`, `fal` | the model named in capabilities | | `durationSeconds` | model-specific | seedance/kling ~5; **hailuo 6 or 10** (1080p → 6 only); **grok 1–15**; **gemini-omni 3–15 (first+last 3–10; omit for 10)**; **ofox 2–30 (per-model API limits)**; **agnes 4–12 (default 5)** | | `ratio` | see model docs | 16:9 (seedance/kling/grok/ofox); **gemini-omni 16:9\|9:16 only**; **ignored on hailuo** | | `resolution` | `360p`, `480p`, `512p`, `720p`, `1080p`, `4k` | provider-specific; hailuo adds 512p for Hailuo-02; grok: 480p/720p/1080p; gemini-omni: 360p/720p/1080p/4k | | `refVideoMode` | `feature`, `base` | kling only, with `refVideos` | | `promptOptimizer` / `fastPretreatment` | boolean | hailuo only | | `generateAudio`, `seed`, `cameraFixed`, `watermark` | controls | seedance only | | `returnLastFrame`, `executionExpiresAfter`, `priority` | controls | seedance only; requested last frame becomes another image asset | | `name` | descriptive asset name | required for good pool UX | | `firstFrame` | project image asset ref | optional | | `lastFrame` | project image asset ref | seedance / kling / hailuo (requires firstFrame; not with multi-ref on seedance) | | `refImages` / `refVideos` / `refAudios` | asset refs | seedance full; kling: images + **1** feature video (no audio); hailuo: none (frames / S2V subject) | | `mode` / `shotType` / `multiPrompts` | Kling multi-shot | kling only |

Model-specific params — see the model's reference.

Input Resolution

`firstFrame` / `lastFrame` / `refImages` / `refVideos` / `refAudios` all take a project asset reference. Prefer a full UUID or short prefix from `read_project`; `asset://<id>` and same-project asset URLs retur