scripts

ALWAYS follow this style when writing Python / JavaScript code

To select, design, annotate, and verify a data visualization (chart, map, or table).

ALWAYS follow this design guide for any front-end work

Use this for non-trivial analysis, design, diagnosis, review, strategy, or judgment. Skip it for simple lookups, mechanical rewrites, and pure tone tasks.

For CloudFlare development, deployment, e.g. Python CloudFlare Workers

Use when creating demos or POCs

Navigate LinkedIn via Chrome DevTools Protocol (CDP) using `uvx rodney` connected to an existing logged-in browser session. Use for extracting profile information, discovering people by city/role/interest, finding people who can help with specific tasks, and surfacing interesting posts and content.

Workflows for melt, the MLT Framework CLI video editor, to edit, transcode, composite, stream, or automate video/audio. Covers multitrack composition, color grading, batch processing, encoding, streaming, and server-side pipelines.

Conventions for package.json, README.md, coding & testing styles

How to plan & break down large, complex tasks

Tips on using uv and uvx (Python build tools) effectively with GitHub, Torch, etc.

Vector art assets (characters, objects, scenes) sources for SVG/Canvas and how to animate them

Use vitest + jsdom for fast, lightweight unit tests for front-end apps

Use to create tooltips, popups, interactions, and animations as informative and engaging aids.

Call LLM via CLI for transcription, vision, speech/image generation, piping prompts, sub-agents, ...

If any tool call fails, log it IMMEDIATELY with this skill

Use when writing command-line scripts agents will control

Use to investigate data for surprising, actionable insights

Write data findings as a compelling narrative story, Malcolm Gladwell prose, NYT graphics-team visuals, engaging & memorable even for a non-technical audience.

Use CDP at localhost:9222 to test/debug websites, automate browser tasks

If the task had failures or was inefficient or complex, ALWAYS run this post-mortem.

Use interactivity purposefully in data visualizations and narrative content. Trigger for scrollytelling, explorable explanations, simulations, guess-then-reveal patterns, or any data story where user action shapes understanding. NOT for dashboard drill-downs or generic UI interactivity.

How to read, manipulate, and generate PDFs using CLI tools and Python libraries.