Skip to main content
skill CX Codex

run-longcat-avatar-video — Claude Code skill: how to install

run-longcat-avatar-video is a Claude Code skill that orchestrates talking avatar generation via draw-things-cli. It animates a portrait from sound input in LongCat-Video-Avatar 1.5.

GPL-3.0 ★ 171 updated 2026-11-17

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add drawthingsai/local-code-community --skill run-longcat-avatar-video -a claude-code
Open on GitHub

What run-longcat-avatar-video does

This skill prepares CLI execution pipelines for the LongCat-Video-Avatar 1.5 diffusion setup. It sets distillation values to 8 steps, CFG 1, and shift 7 at 25 frames per second. Clip lengths follow the strict 4k + 1 requirement. A standard standalone generation uses 93 frames. When processing extended speech recordings, it activates local AVC mode with 93 segment frames and 13 conditioning frames, yielding 80 unique frames per step.

Running the tool requires compiling draw-things-cli with Bazel inside the Draw Things source tree. The local models directory must contain the LongCat checkpoint (i8x or q8p), UMT5 text weights, the Wan 2.1 video VAE, and whisper_large_v3_f16.ckpt. Processing halts immediately if the Whisper checkpoint is absent because LongCat does not fetch it automatically. Audio inputs are pre-validated through afinfo to confirm valid duration and container decoding.

What it helps with

  • render animated avatar videos from a still image and an audio file
  • configure local LongCat-Video-Avatar 1.5 pipelines on Apple silicon
  • process long speech files into segmented AVC chunks using 93/13 framing
  • validate audio container integrity via afinfo before running diffusion

Example prompts for the agent

  • Generate an avatar clip using portrait.png and voice.wav with the LongCat i8x model
  • Render an extended video using local AVC mode 93/13 for speech.mp3
  • Check the local Whisper checkpoint and build a 93-frame clip without AVC

How to install run-longcat-avatar-video

1Skills CLI (Vercel)

npx skills add drawthingsai/local-code-community --skill run-longcat-avatar-video -a claude-code

2Manually: put the files in place

~/.claude/skills/run-longcat-avatar-video/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/run-longcat-avatar-video/

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

Questions about run-longcat-avatar-video

How to install run-longcat-avatar-video in Claude Code?

Run npx skills add drawthingsai/local-code-community --skill run-longcat-avatar-video -a claude-code. You can also place the manifest directly at ~/.claude/skills/run-longcat-avatar-video/SKILL.md.

What does run-longcat-avatar-video do?

It orchestrates command line flags for Draw Things to create talking portrait videos. It enforces dimensions divisible by 64, sets distillation steps, and manages audio conditioning.

What is required to run run-longcat-avatar-video?

You need a draw-things-cli binary compiled with Bazel, macOS afinfo, and downloaded weights. The model path requires LongCat checkpoints, UMT5, Wan VAE, and whisper_large_v3_f16.ckpt.

Similar skills

All skills for AI and ML →

From the same set20

The whole set →
deploy-static-site-with-wrangler deploy-static-site-with-wrangler is a Claude Code skill that uploads static site builds to temporary public Cloudflare URLs. It triggers when someone asks to share or preview built HTML assets. CX Codex +1 ★ 171repo stars train-lora train-lora is a Claude Code skill that validates LoRA training with Draw Things through draw-things-cli. It runs local dataset checks and compares generated images with and without the LoRA. CX Codex +1 ★ 171repo stars skill-creator skill-creator is a Claude Code skill that guides creating, validating, updating, and removing workspace skills. It triggers on requests for reusable workflows or tool integrations. CX Codex +1 ★ 171repo stars init-gpu-server init-gpu-server is a Claude Code skill that prepares a Draw Things GPU server by syncing GPUScript, installing server software, and mounting data disks. CX Codex +1 ★ 171repo stars trainer-integration trainer-integration is a Claude Code skill for adding or refining LoRA training support for models in Draw Things. It covers trainer wiring, checkpoint export, and validation. CX Codex +1 ★ 171repo stars use-drawthings-cli use-drawthings-cli is a Claude Code skill for running and troubleshooting image generation with Draw Things CLI. It comes into play when choosing a model, setting up cloud access, or diagnosing a failed run. CX Codex +1 ★ 171repo stars setup-cpu-proxy-server setup-cpu-proxy-server is a Claude Code skill that deploys Draw Things CPU proxy and Envoy instances on Linux hosts. Runs during node provisioning via SSH or Tailscale. CX Codex +1 ★ 171repo stars new-model-integration new-model-integration is a Claude Code skill for adding image or video generation models to Draw Things. It guides the work from compilation through model wiring and CLI checks. CX Codex +1 ★ 171repo stars

Project sponsors

Companies that keep this analytics open