
Caricature Roast
Stage your own comedy dialogue as a cartoon conversation with a microphone guest, a reacting host and speaker-specific captions.
- Included with SkillGild Pro
- Runs locally in your agent
- v1.0.1
- SkillGild
skillgild install caricature-roastLicense: SEE LICENSE IN LICENSE · Created by SkillGild

What you get
- Stage your own comedy dialogue as a cartoon conversation with a microphone guest, a reacting host and speaker-specific captions.
- A split-stage conversation layout
- Speaker-specific mouth animation and reactions
- Punchline cuts and Hindi or English captions
Is it a good fit?
Good for
- Creators making cartoon & anime stories with a coding agent.
- A project starting from a show clip (MP3) with the guest and host talking.
Not for
- A hosted rendering service or a guaranteed one-click edit; the workflow runs in your coding agent.
- Media you do not have rights to use, or recreating a public figure or another show's branding.
Requirements
- A coding agent that can read local files and run commands, with access to your own or licensed input media.
- Node.js 22+ and npm for HyperFrames 0.8.193; ffmpeg/ffprobe for local media processing and MP4 export.
- Python 3 plus Whisper CLI and a downloaded speech model when deriving timings from audio. Supply a corrected transcript for captions.
- numpy and librosa for the bundled amplitude-based mouth helper when a character speaks. Separate voices or speaker labels are needed for multi-speaker scenes.
- Speaker-labelled dialogue or a separately installed diarization tool. Supply separate voice tracks for reliable per-speaker mouth timing.
Install the Caricature Roast skill
Choose your agent and run one command. SkillGild signs you in when needed and connects the skill for you.
Install the SkillGild CLI first. If you already have it, run skillgild update once.
Install this skill for Claude Code.
Terminal skillgild install caricature-roast --agent claude-codeOn first use, approve the sign-in request in your browser. SkillGild connects Claude Code automatically. Restart or reload your agent, then ask it to use this skill.
Developer detailsAPI examples, server tools and input schemas
How it runs
- Your agent starts a session. SkillGild checks your access and allowance, counts one run, and sends the skill's instructions for this task.
- Your agent does the work with its own model and tools, in your project. SkillGild makes no AI model call.
Use it from code
Start a session, follow its guide, then end it. Create an API key under Account and pass it as a bearer token.
# Prints the session ID, its tools and the instructions for your agent
skillgild session caricature-roast
skillgild end <session-id>Questions
What does Caricature Roast do?
Stage your own comedy dialogue as a cartoon conversation with a microphone guest, a reacting host and speaker-specific captions.
What input does Caricature Roast need?
a show clip (MP3) with the guest and host talking. Speaker-labelled dialogue or a diarization tool plus separate voice tracks for reliable per-speaker lip-sync; no speaker-label model is bundled.
Do I need to install another video skill?
No. Run skillgild install caricature-roast --agent claude-code, or choose codex, cursor or gemini-cli. Each listing includes its own shared engine resources; no sibling skill is required. Your agent then follows this skill's session guide.
Where does the work run?
In your coding agent on your computer, using local rendering and transcription tools. The package includes a template and timing helpers. Extra models and specialist tools in Requirements are supplied separately.
Is Caricature Roast Free or Pro?
Included with SkillGild Pro, using your plan's session allowance. Local model, rendering and any external service costs are separate.
What needs checking before using Caricature Roast?
Node.js 22+ and npm for HyperFrames 0.8.193; ffmpeg/ffprobe for local media processing and MP4 export. Python 3 plus Whisper CLI and a downloaded speech model when deriving timings from audio. Supply a corrected transcript for captions. numpy and librosa for the bundled amplitude-based mouth helper when a character speaks. Separate voices or speaker labels are needed for multi-speaker scenes. Speaker-labelled dialogue or a separately installed diarization tool. Supply separate voice tracks for reliable per-speaker mouth timing.
Related guides
Setup steps and worked examples for this kind of skill.
- Claude Code video skills: how code-rendered video worksCompare Brag, HyperFrames, fframes, Lyric Music Video and Video Illustrator, and see what a code-rendered video workflow needs on your machine.
- How to install Claude Code skills (and skills for Codex, Cursor and Gemini CLI)Install a Claude Code skill from GitHub, a plugin or SkillGild in one command. Folder paths, personal vs project skills, verifying it loaded, troubleshooting, and the same steps for Codex, Cursor and Gemini CLI.
- How to create a Claude skillCreate a Claude skill from scratch: write SKILL.md with a name and description, add scripts and references, test it in Claude Code, make it work in Codex and Cursor, and share it on GitHub, as a plugin or on SkillGild.
Reviews
No reviews yet
Used Caricature Roast? Your rating helps the next person decide.





