> For the complete documentation index, see [llms.txt](https://animotive.gitbook.io/animotive-kimodo/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://animotive.gitbook.io/animotive-kimodo/working-in-sequencer/keying-poses.md).

# Keying Poses

Animotive Kimodo generates motion by interpolating between poses you set on the character's Control Rig. Each Control Rig keyframe is one input pose; each gap between two adjacent keyframes is one in-between window for the generator.

> **Control Rig** is Unreal Engine's in-editor rigging surface — the layer that exposes animator-facing controllers (coloured viewport handles) on top of a skeletal mesh. For a MetaHuman, the body skeletal mesh ships with a Control Rig already attached. For other characters, the Retarget Wizard can auto-generate one (see [First-Time Character Setup](/animotive-kimodo/setup/first-character.md)).

## What you're keyframing

* The **frame** of each Control Rig keyframe defines the time boundary of an in-between window.
* The **pose** sampled at that keyframe becomes a hard constraint at that window's start or end.
* The Animotive Kimodo prompt section spanning the gap between two adjacent keyframes drives what motion fills that gap.

Without Control Rig keyframes, only the text-to-anim path is available — a single prompt with no spatial constraints. Most authored work uses two or more key poses to lock the motion's shape.

## The keying loop

1. Open the Level Sequence containing your character in Sequencer.
2. Expand the character's body sub-binding (typically `Body`) and locate its **Control Rig** section.
3. Scrub the playhead to the frame for your first pose.
4. Click the character in the viewport to reveal the Control Rig handles. Manipulate them into the pose.
5. With the viewport focused, press **S** (Set Key).
6. Scrub forward to the next pose frame, manipulate, press **S** again.

Repeat for as many key poses as the clip needs. Two key poses produce one in-between gap; three produce two adjacent gaps, and so on.

## Common pitfalls

* **Posing the face rig.** Animotive Kimodo only consumes body motion. Keyframes set on a MetaHuman face Control Rig (the one attached to the `Face` sub-binding) are filtered out by the exporter. If a pose change is not showing up in generations, confirm you are keying the body rig.
* **Posing in the wrong reference space.** Translating a Control Rig handle in world space when its default space is parent or local space can drift the rig in unintuitive ways across keyframes. Leave handles in their default reference space unless you have a specific reason to switch.
* **Section row vs Control Rig row.** **Add Key** is meaningful only when invoked on the Control Rig section row. Invoking it on the parent binding row keys nothing useful.

## See also

* [Animotive Kimodo Track](/animotive-kimodo/working-in-sequencer/animotive-kimodo-track.md) — prompt and section authoring on top of your key poses.
* [Constraints Track](/animotive-kimodo/working-in-sequencer/constraints-track.md) — companion track for pinning body regions on specific frames.
* [Prompting with Animotive Kimodo](/animotive-kimodo/getting-started/prompting-with-animotive-kimodo.md) — the prompting walkthrough this page expands on.
* Epic Games documentation: [Animating with Control Rig in Unreal Engine](https://dev.epicgames.com/documentation/en-us/unreal-engine/animating-with-control-rig-in-unreal-engine) — authoritative Control Rig reference.