Text to motion skill for AI agents

​

When your agent uses it

• A custom action the curated motion library lacks — nod, bow, shrug, wave with a specific arm, an emotional gesture
• Animating a non-humanoid rig — a talking flower, animal, creature, turret or prop
• Adding clips to an uploaded or custom-rigged GLB/FBX model
• “make the flower nod” / “animate my creature from a text description”

​

Install

npx -y summer-engine@latest skills install text-to-motion --agent claude-code

​

The skill

​

Choose the right route

​

The order

1. Confirm the rig. The asset must be the user’s own 3d_model GLB/FBX with one skinned armature and 5-70 bones (assets.search({ scope: "mine", assetType: "3d_model" }) / assets.get). If it is unrigged or has odd bone names, read rigging-for-text-motion.md in this directory.
2. Agree the clip list and cost. Each prompt x take is one billed clip. Run the call below with dryRun: true first: it checks the rig and returns estimate.credits. Show the list and the credits and wait for the user’s OK before submitting.
3. Submit once. Up to 8 prompts per call, takes 1-4. For rooted characters pass lockJoints with the base bones:

   assets.generate_motion({
     rigAssetId: "<id>",
     backend: "text-to-motion",
     prompts: ["nods happily", "waves hello with the right leaf"],
     lockJoints: ["Hips", "Spine"],   // rooted flower; omit for characters that move
     cfgScale: 5                      // default 3; 5 for clearer gestures on stylized rigs
   })

​

Hard rules

• Only the user’s own rigged models. Never animate assets owned by someone else.
• Clips are 2 seconds and not guaranteed to loop. For longer actions chain several prompts (“raises both arms”, “holds the pose”, “lowers the arms”) or loop a good take.
• Locomotion moves the root. Walk/run/jump clips may translate the character; gameplay code owns movement, so strip or ignore root translation for in-place controllers.
• Never lock joints on characters that walk or jump; the root would be pinned.
• Report failures with their code. Job failures starting text_motion: (no skinned armature, too many or too few joints, multiple armatures, unknown lock joint) are input problems and the charge is refunded. Fix the rig or the request; do not retry unchanged.
• A backend without text-to-motion means it is not enabled on this server yet. Fall back to generate-motion (meshy-library) for humanoid clips and say so; do not retry.
• Do not promise facial animation, finger detail or lip sync. The model animates the skeleton only; blendshapes stay untouched (facial-and-lipsync covers faces).

​

Request errors

​

Without Summer Engine

​

Do not stop early

​

From a shell

​

Files in this skill

• prompts-and-controls.md
• rigging-for-text-motion.md

​

Related skills

• generate-motion: Attach a library animation clip (idle, walk, run, attack) to a rigged character via AnimationPlayer.
• character-model: Generate a rigged humanoid (player, NPC, enemy, boss) from a T-pose reference, previewed before rigging.
• retarget: Apply animation clips from one rigged character to another: one clip library, many models.
• asset-strategy: Route any ‘I need a [thing]’ asset request to the right skill: 2D, 3D, audio, video, VFX or animation.