- Home
- Skills
- Workflow Automation
- Stickers
Stickers
Your own character as a WhatsApp and Telegram sticker pack: an open recipe, a code cutout and a folder ready to import.
Free
Works with the AI Agents you already use
Stickers
Example session with this skill installed
Make a WhatsApp sticker pack of my character Mika. Here are 3 reference images. Twelve moods, and put "NOPE" on the annoyed one.
- Read your context and instructions
- Compiled the stickers
Mika's pack is ready: 12 stickers in one folder, ready to import.
| Step | What happened |
|---|---|
| References | Your 3 images went with every generation, and Mika's fixed traits filled the prompt's first slot: silver bob, red eyes, yellow raincoat. |
| Generation | 12 images on flat #00E000, named 01-laughing.png to 12-whatever.png. 07-sleeping came back with a gradient behind the head and was generated again. |
| Cutout | node tools/cutout.mjs --in mika/input --out mika/output: 12 transparent 512x512 WebP, all under 100KB. 05-crying kept a green pocket between the arm and the body, cleared with --seeds 50. |
| Caption | node tools/caption.mjs mika/output/04-annoyed.webp "NOPE" --out mika/output-captioned. Drawn in code, so the letters come out clean. The plain version stays in mika/output. |
| Pack | The captioned 04 plus the other 11 went into mika/final, then node tools/pack.mjs --dir mika/final --out mika/packs --name "Mika" --author "You". The placeholder emojis in contents.json were swapped for the ones that match each mood: 😂 😭 😴 😒 and the rest. |
Install
- WhatsApp: Sticker Maker (Android) or Sticker Maker Studio (iOS) imports the folder in
mika/packs/in one go. - Telegram: BotFather, then @Stickers, then /newpack, and upload the 12 files.
Next steps
- Open two or three stickers on your phone before sharing: the white border should close all the way around.
- Want more moods? Name them and the same cycle runs again. Past 30, WhatsApp gets the pack in parts, one folder each.
Connects securely to your tools. The creator never sees your data.
About this skill
Manual sticker creation for WhatsApp and Telegram is tedious, involving inconsistent character designs, complex background removal, and strict platform formatting. This skill provides a structured, code-driven pipeline to transform character references into professional sticker packs. It automates the generation, cutout, and assembly process while maintaining character consistency across dozens of expressions.
What it does
- Character Consistency. Uses 1-4 reference images to ensure a character maintains fixed traits across the whole pack.
- Automated Background Removal. Executes a code-based cutout using a chroma-key method (#00E000) to create transparent WebP files without needing segmentation models.
- Dynamic Captioning. Overlays legible text onto generated stickers via code, avoiding the common spelling errors found in AI image text.
- Platform Optimization. Automatically resizes and compresses images to 512x512 WebP files under 100KB to meet WhatsApp and Telegram technical requirements.
- Pack Organization. Groups stickers into folders of 30 for WhatsApp compatibility and generates the necessary
contents.jsonmetadata.
How it works
- Define Character. You provide 1-4 reference images and a list of expressions (e.g., laughing, sleeping).
- Generate Assets. The skill provides a specific "recipe" prompt for your image generator to produce stickers with a flat green background.
- Execute Cutout. Run
node tools/cutout.mjsto remove backgrounds and normalize the images. - Apply Captions. Optionally use
node tools/caption.mjsto add text overlays to the finalized images. - Assemble Pack. Run
node tools/pack.mjsto generate the installable folder structure and metadata.
Frameworks & tools
This skill requires Node.js 18+ and the sharp image processing library. It uses custom MJS scripts for flood-fill background removal and JSON manifest generation.
Why this beats prompting it yourself
Generic prompts often fail to produce the mandatory margin needed for sticker borders or result in "hallucinated" text that is unreadable. This skill uses a proven two-pass cutout logic that handles enclosed gaps (like the space between a character's arms) which simple background removers often miss.
Use cases
- OC to Stickers. Converting an original character into a full reaction pack for personal or community use.
- Brand Assets. Creating a standardized set of mascot stickers for a company Telegram channel.
- Batch Processing. Turning a folder of raw AI generations into a formatted, transparent, and captioned sticker set.
Known limitations
Requires Node.js 18 and a local installation of the sharp dependency. Characters with green hair or clothing matching #00E000 may require manual tolerance adjustments in the cutout script.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.
Needs Node.js 18+ and sharp (npm install once inside the kit). Generation runs on your own image generator, which must accept reference images. A character dressed in chroma green needs the tighter cutout flags (--seeds 30 --tolerance 45).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 27 days ago
- Free with an Agensi account
- Software
Node.js 18+sharpImage generator with reference images
sharp installs from npm once inside the kit.
Permissions
Terminal: runs the three Node scripts, and npm install (sharp) once inside the kit. Read and write files: reads the approved images and writes the cutouts, captioned versions and pack folders in the person's own work folder, never inside the kit. Network: the npm install only.