Update to Visual Sandbox

{
  "description": "Turn one photo into a short video clip with the Visual Sandbox MCP tools: a product shot that moves, a portrait that comes alive, a still that gets a slow camera move. Trigger when the person asks to \"animate this photo\", \"make this picture move\", \"turn my product photo into a video\", \"add motion\", \"image to video\", or attaches a picture and asks for a clip. Keeps the subject unchanged, asks for the price first, and shows the clip in a card. Uses the MCP tools only.\n",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 280
    }
  ],
  "name": "vsb-animate-photo",
  "skill_md_contents": "---\nname: vsb-animate-photo\ndescription: >\n  Turn one photo into a short video clip with the Visual Sandbox MCP tools:\n  a product shot that moves, a portrait that comes alive, a still that gets\n  a slow camera move. Trigger when the person asks to \"animate this photo\",\n  \"make this picture move\", \"turn my product photo into a video\", \"add\n  motion\", \"image to video\", or attaches a picture and asks for a clip.\n  Keeps the subject unchanged, asks for the price first, and shows the clip\n  in a card. Uses the MCP tools only.\nlicense: Proprietary. See LICENSE for the terms.\n---\n\n# Animate a photo\n\nOne photo becomes one clip of about 5 seconds. The subject stays the same.\nOnly the motion is new. Use the MCP tools only.\n\n## 1. Get the photo and the motion\n\n- A photo attached in the chat goes to `upload_media` as `file`. Use the\n  `url` it returns. A public `https` link can go in directly.\n- Ask what moves, in one question, only if the person did not say. Offer\n  two choices: a slow camera move, or one small action of the subject.\n- Keep the shape of the photo. A tall photo makes a 9:16 clip, a wide photo\n  makes a 16:9 clip.\n\n## 2. Pick the model\n\nRead [`vsb-video`](../vsb-video/SKILL.md) with `get_skill`, at \"Picking a\nmodel\". Then use this short rule:\n\n- **A first try, or a product shot:** `video/p-video` or `video/p-video-2`\n  with the photo as `image`. Fast and low in cost.\n- **People, hands, or a hard motion:** `video/seedance-2.0-mini` with the\n  photo as `image`. It holds faces and hands better.\n- **Speech:** this is not the right skill. Read `vsb-talking-avatar`.\n\n## 3. Write the prompt\n\nDescribe only the motion and the camera. The photo already holds the look.\n\n- Name one motion: \"steam rises from the cup\", \"she turns her head and\n  smiles\", \"the camera pushes in slowly\".\n- Add what must not change: \"The product keeps its exact shape, colour and\n  label.\"\n- Leave out new objects, new text and scene changes. They make the model\n  redraw the photo.\n\n## 4. Price, run, show\n\n1. Call `estimate_cost` with the model and the inputs. Tell the person the\n   price and wait for their yes.\n2. Run the model with the photo `url`, the prompt, the aspect ratio and a\n   duration of 5. Name the work with `sandbox_name`. In Review mode, a run\n   of 10 cents or more waits for the person's press of Generate: show the\n   card with `show_media`, or give the review link. Never press it for them.\n3. Call `show_media` with the job id at once. The card shows the percent and\n   then plays the clip.\n4. When the person wants to judge it yourself, call `get_job` after\n   `eta_seconds`. It returns four frames of the clip.\n\n## Rules\n\n- One clip per request unless the person asks for more.\n- When the label or the face changes, say so. Offer a second run with a\n  smaller motion, not a longer clip.\n- To make it vertical later, or to add music, read `vsb-edit-video`.\n"
}