# Titles and motion

> Add titles by hand or from the chat in nolanlabs, have the agent make motion graphics on the timeline, and reuse them with templates and references.

Page: https://nolanlabs.io/en-gb/docs/motion

Words and motion graphics live on the timeline with your cut, so you make them where you cut. A title you can set by hand. A motion graphic you ask for in the chat.

## Add a title

**In Pro**, press the T button in the timeline toolbar. It adds a title at the playhead. Set its words, size, height and color in the Inspector. **Style** applies a look, picks the face, and sets how the words come in, go out and move while they hold.

**In Lite**, the **Text** tab puts the words at the playhead for three seconds, in the look you picked. Pick them on the timeline to type over them, and the picture follows the typing.

The looks: bold caption, karaoke, headline, impact, sport, outline, poster, kicker, boxed, lower third, clean, cinematic, elegant, neon, comic, marker and mono.

From a script or the agent, words can also run along a curved line, and letters can come in one after another.

## Faces

A title can use one of the 28 faces that come with the editor, or a TrueType or OpenType file from the project’s `Fonts` folder or your computer. Importing a font copies it into the project, so it travels with it. The project’s own faces win over the computer’s.

A title naming a face that isn’t here fails its check and falls back, so a missing font is reported, not silent.

## Ask for a motion graphic

A motion clip is a motion graphic written as a small web page, drawn one frame at a time and laid on the timeline like a title. Ask in the chat, for example: “add a lower third for Ada Park”.

You can then trim, move, fade, blend and transform it like any clip, and key its params. It survives proposals and undo.

The agent can start from these, or write one whole:

- **Words:** lower third, kinetic title, title card, end card, captions, callout.
- **Data:** stat, chart build, feature list.
- **Pictures:** wipe, glow field, 3D product turn, world with host, layered parallax, 3D cards.

A clip can read other clips on the timeline, each frame with its alpha, so a page can put your footage on a turning screen. A clip can use GSAP, Three.js, anime.js, Lottie and D3, which come with the editor, and nothing is fetched from the network.

The motion check looks for what can go wrong: frames that draw differently a second time, missing files, words outside the title-safe area or cut off by their box, faces not found and low contrast.

The drawing browser downloads on first use. Where that can’t happen, your computer’s own Chrome or Chromium draws. Frames are kept, so each change draws once, and what plays is what renders. Clips draw at 1080p on the short side; Pro can draw at 4K.

## Use it elsewhere

A motion clip can be written on its own with its alpha: ProRes 4444, WebM (VP9) or PNG frames, at 1080p or 4K in Pro. Only OTIO carries motion clips between apps. See [Hand off](https://nolanlabs.io/en-gb/docs/hand-off).

## Templates

A template is a section of a timeline kept to lay down again with new material: its clips, motion clips, titles, looks and timing, with slots for what changes. Ask the agent to lay one down. These are built in:

| Template | What it is |
| --- | --- |
| Documentary interview | A chapter card, then an interview with a lower third |
| Film title sequence | A credit and a title over a slow push into a plate, with grain, halation and a vignette |
| Tutorial callouts | Two rings with labels over a screen recording, and a caption |
| Product launch | A device turning with a screen on it, the features, an end card |
| Social vertical | Three shots with a pop on each cut, captions, an end card, for 9:16 |

To keep your own from a script, use `template.save`. See [Scripting](https://nolanlabs.io/en-gb/docs/scripting).

## Follow a reference

The agent can read how a finished piece is cut: its shots and lengths, cuts against the music’s beats and hits, palette, words on screen and movement per shot, each shot with its reason. It’s signal processing, and it finds where words are, not what they say.

## Limits

Cut-outs come only from a picture’s own alpha or a keyed green screen. There’s no camera tracking to fix a graphic into a moving shot. See [Known limits](https://nolanlabs.io/en-gb/docs/known-limits).
