AI Photo Animator MCP server

• Registry nameai.aiphotoanimator/photo-animatorverified in the MCP registry

The registry name sits under aiphotoanimator.ai. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

• Stateanswerschecked 2026-10-29
• Tools10
• Server calls itselfai-photo-animator
• Server version1.0.0
• Protocol version2025-06-18

On 2026-10-29 the server answered initialize and tools/list without a sign-in. First answer in 115 ms.

• Its tools do every use case we checked: animate old photos into short videos with sound; make pet photos move with one-click templates; turn drawings and paintings into animated clips.
• Checked on 5 of the 7 days it takes to judge reliability.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

• INITE MCP scoreafter a weeka score needs 7 daily checks: 5 so far
• What an agent can do with it3 of 3 use cases
• Answeredmeasuring5 of the 7 daily checks it takes to print a share
• Median first answer153 ms
• Space in an agent's context≈ 1,324 tokensthe definitions of all its tools (10), which an agent carries on every turn
• Tool definitionsA · 98 of 100
• Typed results0 of 10 toolsdeclared outputSchema: results come typed, read once

• Parameters without a description: in 1 of 10

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

• list_effects

  List video effect templates (dance, sports, meme scenes, etc.). Use the returned `effect` id with animate_photo. `photos` is how many photos the effect needs.

• animate_photo

  Turn photo(s) into a short video. With `effect` (from list_effects) the effect decides the motion; without it the photo is animated following `prompt`. Costs credits. Returns a task_id; poll get_video for the result (usually 1–5 minutes).

• create_video

  Generate a short video (5 seconds, 720p, with sound) from a text description only, no photo needed. Costs credits. Returns a task_id; poll get_video.

• edit_video

  Edit an existing video clip (MP4, up to 15 seconds, 50 MB) following instructions: put the person from a photo into the clip, change outfit, style, background or scene, while keeping the motion and timing. Costs credits (based on clip…

• create_image

  Generate a new image from text, or edit photo(s) following instructions (change background, outfit, style, lighting, combine several photos). Uses Nano Banana 2 at 1K. Costs credits. Usually returns the image directly; if status is…

• restore_photo

  Repair an old or damaged photo: remove scratches, dust, creases and fading, recover sharp faces, keep everyone exactly as they are. Optionally colorize a black-and-white photo. Costs credits. Returns the restored image (or pending →…

• get_image

  Check an image task from create_image or restore_photo. status is pending, success (image_url ready) or failed (credits refunded).

• get_video

  Check a generation task. status is pending, success (video_url ready) or failed (credits refunded).

• list_my_videos

  The user's most recent video generations, newest first.

• get_credits

  Remaining credits and plan of the connected account.