bratification

Generate brat-style text cover images and sticker-like WhatsApp-safe outputs from custom text. Use when the user asks to make a brat or brat-gen style cover,...

Security Scan

The skill's code and instructions align with its stated purpose (rendering brat-style images); it reads/writes local image files, uses local fonts, and does not request credentials or network access.

Purpose & Capability

Name/description describe text-to-image 'brat' rendering and the bundled Python scripts implement exactly that (PNG/WEBP output, !brat command parsing, font handling, sticker-size options). There are no unrelated credentials, binaries, or capabilities requested.

Instruction Scope

SKILL.md instructs running the included scripts and the scripts only read inputs, measure fonts, render images, and write outputs into a managed scripts/out tree. Path sanitization is implemented; the code does not read unrelated files, environment variables, or contact external endpoints.

Install Mechanism

No install spec is provided (instruction-only). Code is included in the bundle but there is no network download or external installer declared. This minimizes install-time risk.

Credentials

The skill requires no environment variables or credentials. The only external dependencies are local font files and the Python imaging library (Pillow), which are proportional to rendering images.

Persistence & Privilege

The skill is not marked always:true and uses default autonomous invocation; it does not modify other skills or system-wide settings. It only creates files under its own scripts/out directory.

Assessment

This skill appears coherent and limited to generating images. Before installing, note: (1) it requires Python with the Pillow library and access to system TTF fonts — otherwise rendering will fail; (2) it writes output files under scripts/out (the code sanitizes paths to avoid traversal); and (3) there is no network or credential usage in the bundle. If the source is untrusted, run the scripts in a restricted environment first (or inspect/execute them locally) — but from a security/permission perspective the behavior matches the stated purpose.

Generate brat-style images from text using the bundled Python scripts.

Quick start

• For a plain render, run scripts/render_brat.py.
• For command-style parsing like !brat hello world, run scripts/handle_brat_command.py.
• Prefer white background, Arial Narrow bold (with cross-platform font fallback), centered composition, word wrapping without cutting words, and automatic font shrink to fit.
• handle_brat_command.py now sanitizes slug/output naming and always writes into its managed scripts/out directory tree.

Workflow

1. Decide whether the request is direct text rendering or !brat command parsing.
2. Use scripts/handle_brat_command.py for command-style input.
3. Use scripts/render_brat.py for direct text-to-image rendering.
4. For WhatsApp-safe sticker-like delivery, prefer --size 512 --sticker-size 512 and send the generated PNG with no caption.
5. Treat WEBP as an optional artifact unless the channel has verified native sticker upload support.

Commands

Render text directly

python scripts/render_brat.py "hello world" --png out/hello-world.png --webp out/hello-world.webp

Parse a !brat command

python scripts/handle_brat_command.py "!brat hello world"

Sticker-like WhatsApp-safe output

python scripts/handle_brat_command.py --size 512 --sticker-size 512 "!brat hello world"

Output handling

• Use PNG as the primary reliable delivery format.
• Use WEBP as a secondary artifact for future native-sticker-capable channels.
• If the input is only !brat with no text, ask briefly for the text.
• If generation fails, apologize briefly and say the brat render failed.

Scripts

• scripts/render_brat.py: deterministic renderer for PNG/WEBP output
• scripts/handle_brat_command.py: parser for !brat-style input that emits JSON with generated paths