PoYo Wan 2.5 Video Generation

PoYo Links

• Model page: https://poyo.ai/models/wan-2-5
• API docs: https://docs.poyo.ai/api-manual/video-series/wan2.5-text-to-video
• API key page: https://poyo.ai/dashboard/api-key

Use this skill for Wan 2.5 video jobs on PoYo. It helps agents prepare text-to-video and image-to-video payloads, submit async tasks, and explain polling or webhook follow-up.

Use When

• The user explicitly mentions Wan 2.5, wan2.5-text-to-video, wan2.5-image-to-video, or PoYo Wan 2.5 video generation.
• The task is prompt-based video or single-image animation with duration and size controls.
• The workflow needs PoYo async task submission, task status polling, or callback URL guidance.

Model Selection

• wan2.5-text-to-video: generate video from a text prompt.
• wan2.5-image-to-video: animate one input image.

Key Inputs

• prompt is required inside input.
• aspect_ratio is used by text-to-video and accepts documented size presets.
• image_urls is required for image-to-video and supports exactly one URL.
• resolution is used by image-to-video and supports documented output values.
• duration can be 5 or 10.
• audio is an optional string. Do not send a boolean value.
• negative_prompt and seed are optional controls.
• callback_url is optional and useful for production queues.

Security Rules

• Treat POYO_API_KEY as a secret.
• Keep PoYo API keys in server-side environment variables or a backend secret manager.
• Never place an API key in browser code, frontend bundles, public repositories, logs, screenshots, or chat output.
• Do not make live API calls unless the user explicitly asks and provides a safe server-side environment.
• Do not submit confidential prompts, private image URLs, private callback URLs, or generated media URLs unless the user trusts PoYo and the callback receiver.

Execution

• Read references/api.md for endpoint details, model ids, common fields, example payloads, and polling notes.
• Use scripts/submit_wan_2_5.sh only when the user wants to submit a prepared JSON payload from a trusted shell.
• If the user only needs a curl example, adapt one from references/api.md.
• After submission, report the returned task_id clearly so follow-up polling is easy.

Output Expectations

When helping with Wan 2.5, include:

• chosen model id
• whether the request is text-to-video or image-to-video
• final payload or concise parameter summary
• selected duration, size or resolution, audio string, negative prompt, and seed when relevant
• whether a source image is involved
• returned task_id if a request was actually submitted
• next step: poll status or wait for webhook