LaunchKit · 2026
Back to Skills

Anima Avatar - Interactive Video Generation Engine.

0
1.2k downloads
by @hmyaoyuan

Setup & Installation

openclaw skills install @hmyaoyuan/anima

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install anima

Version History

v3.3.2Feb 12, 2026 - Version note by skill developer:

Restored original package name 'anima' under @HMyaoyuan.

What This Skill Does

Generates 16:9 avatar videos where a character sprite speaks scripted text with emotion-matched expressions and synthesized voice. Combines Fish Audio for TTS, a library of 30+ character sprites auto-selected by emotion tag, and FFmpeg for frame-by-frame video composition. Finished videos can be uploaded directly to Feishu as native media or saved locally in preview mode.

Handles TTS generation, emotion-based sprite selection, parallel frame rendering, audio sync, and video encoding in one pipeline instead of requiring separate tools for each step.

When to use it

  • Sending an animated character announcement to a Feishu group chat
  • Generating a multi-scene explainer video from a JSON script
  • Producing a custom-voiced avatar response for a chatbot reply
  • Batch-creating onboarding video clips with a branded character
  • Previewing a scripted dialogue locally without uploading anywhere

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Generate a video where the character says 'Good morning team!' happily and then says 'Please review the report' with a thinking expression, and send it to Feishu user ou_abc123

AGENT
  1. 1Parses the request into a two-scene script: [{"text":"Good morning team!","emotion":"Happy"},{"text":"Please review the report","emotion":"Think"}]
  2. 2Calls Fish Audio API to synthesize audio for each line using the configured voice model
  3. 3Selects matching sprites from the library based on emotion tags (Happy, Think)
  4. 4Renders video frames using sharp and SVG text overlay, then composes the final MP4 with FFmpeg
  5. 5Uploads the finished video to the specified Feishu user via send_video_pro.js
OUTPUT

A 16:9 MP4 video delivered to the Feishu conversation with correct duration and native playback

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

Fish Audio API key (FISH_AUDIO_KEY) — sign up at fish.audio/dashboard/apiFish Audio voice model reference ID (FISH_AUDIO_REF_ID) — from your Fish Audio accountGoogle Gemini API key (GEMINI_API_KEY) — optional, needed only for sprite generation, get at aistudio.google.com/apikeyFeishu app ID and secret (FEISHU_APP_ID, FEISHU_APP_SECRET) — optional, needed only for Feishu deliveryffmpeg installed on the system