Skip to main content
skill CX Codex

anymove-film — Claude Code skill: what it does, how to install

anymove-film is a Claude Code skill that creates two-camera AI clips for anymove creatures. It checks the split screen, limb colors, and view timing before capture.

Apache-2.0 ★ 653 updated 2026-11-13

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add patch-gg/anymove
/plugin install anymove@anymove
Open on GitHub

What anymove-film does

anymove-film turns a reference frame from runs/<c>/ref/ref.png into a two-camera movement clip. It builds the action prompt from creature.json, sends the frame to a video model through fal, then checks the result before capture. The front view stays on the left and the side view on the right, with both cameras locked off.

Available model names include h3max, h3, h3maxturbo, seedance, and flux3max. The command tools/film.sh <c> roar h3max 1,2 produces variants for two seeds. Checks reject clips where the split screen breaks, limb colors change, body parts appear or disappear, or joints are covered. The skill requires a fal.ai key as FAL_KEY in .env or the environment, plus a creature already configured with anymove-rig and its reference file.

What it helps with

  • create a two-camera clip for a new creature move
  • keep front and side views synchronized
  • reject clips with swapped limb colors
  • catch broken split-screen framing before capture
  • compare video models for one movement

Example prompts for the agent

  • Create a roar for the creature in assets/<c> and write the prompt to prompts/<c>/roar.txt.
  • Run tools/film.sh <c> roar h3max 1,2 and compare both seeds.
  • Check the clip for a broken split screen, changed limb colors, or disagreement between the views.

How to install anymove-film

1Plugin marketplace

/plugin marketplace add patch-gg/anymove
/plugin install anymove@anymove

This installs the whole anymove plugin with all its skills

2Skills CLI (Vercel)

npx skills add patch-gg/anymove --skill anymove-film -a claude-code

3Manually: put the files in place

~/.claude/skills/anymove-film/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/anymove-film/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about anymove-film

How do I install anymove-film in Claude Code?

Run /plugin marketplace add patch-gg/anymove, then run /plugin install anymove@anymove in Claude Code.

What does anymove-film do?

It writes a movement prompt from creature.json, sends the reference frame to a video model through fal, and checks that the two views retain their split screen, limb colors, and shared timing.

What does anymove-film need to work?

It needs a fal.ai key in FAL_KEY, stored in .env or the environment, plus a creature configured with anymove-rig and a reference file at runs/<c>/ref/ref.png.

Source text

SKILL.md · Apache-2.0
---
name: anymove-film
description: Make the two-camera AI clip for an anymove creature - write the move's prompt from creature.json, send the reference frame to MiniMax H3 Max (or H3, Seedance, FLUX 3 Max) on fal, and check the clip kept the split screen, the limb colours and both views in step before capture. Use when someone wants a new move for a rigged anymove creature, when clips come back with swapped colours, a moving camera or a broken split screen, or when choosing a video model for two-camera clips.
license: Apache-2.0
compatibility: Needs a fal.ai key (FAL_KEY in .env or the environment) and a creature already set up with anymove-rig (runs/<c>/ref/ref.png).
---

# anymove-film: one move, two cameras, one clip

The video model gets one image, `runs/<c>/ref/ref.png` (front view left, side view right, limbs painted), and a prompt
that pins everything except the action. Both views come from one generation, which is what keeps them in step.

## 1. Write the prompt

```bash
.venv/bin/python tools/make_prompt.py assets/<c> "the dragon rears up on its hind legs, spreads its wings wide and roars, then drops back onto all four legs." prompts/<c>/roar.txt
```

The template says: animate the exact model shown twice, front camera left and side camera right, locked off, same
moment frame for frame; keep the model as it is (`describe`); limbs colour-coded and never swapping (`colour_code`);
the body count (`count`); then the action; it stays on the same spot, centred in both views; one continuous 5 s shot
at natural speed; the cameras never move.

Write the action as one sentence that starts and ends in the rest pose ("..., then drops back onto all four legs").
That keeps the creature in frame and makes clips loop. Name the limbs the way `colour_code` does. Keep it on the spot:
walks and runs work as "on the spot like a video-game running loop". Add bans to `count` when the model drifts
("No flames, no smoke, no weapon." for the fire demon). `python tools/make_prompt.py --check` confirms every prompt
under `prompts/` still matches its creature.json.

## 2. Generate

Say the cost first: about $0.26 per 5 s clip with H3 Max (October 2026), one clip per model and seed, and about 7 clips
($1.80) for a usable move on average. Two seeds is a start, not a budget: 58% of captured clips passed the checks.

```bash
tools/film.sh <c> roar h3max 1,2      # runs/<c>/videos/roar/s1/h3max.mp4 and s2/; about 1-2 min each
```

Models (`mocap/generate.py`): `h3max` (MiniMax H3 Max, best sync of the views so far), `h3` (MiniMax H3, $0.30),
`h3maxturbo`, `seedance` (ByteDance Seedance 2.5, expensive: three 720p clips cost $7.10), `flux3max` (FLUX 3 Max). Two
seeds per move is a good default; capture both and keep the one whose views agree better.

## 3. Check the clip before capture

Look at a few frames (or the whole clip) and reject it if:

- the split screen breaks: a view zooms, pans or cuts, the divider moves, the creature walks out of a view;
- a limb's colour changes or swaps sides (plain H3 swapped them on 2 of 5 wolf clips; the reader then labels the
  wrong legs); the LoRA (anymove-lora) kept them on 5 of 5;
- the body count changes (an extra leg, a lost wing), or the creature turns to face another way;
- fire, smoke, dust or a weapon covers joints.

The real gate is numeric and comes in capture: the paw agreement between the views (anymove-capture). What the
clips taught so far:

- Views disagree in the legs, not in time. Seedance's trot played the front legs in anti-phase between the views
  (paw agreement -0.04) while H3 Max scored 0.94 on the same prompt.
- Big single gestures (rear up, roar, slam, take off, pounce) read best: dragon take-off 1.00, spider rear-up 0.99,
  demon wing beat 0.99. Eight-legged walking fails (0.36-0.49): the views disagree on which legs are up.
- Lying down and dying poses failed on the wolf: the reader's random training poses don't cover them.

© Patch · Apache-2.0 · patch-gg/anymove

Similar skills

All skills for AI and ML →

From the same set7

Project sponsors

Companies that keep this analytics open