All skills

Builds a full Amazon PPC plan — SP/SB/SD campaign architecture, budget allocation, and ACoS optimization — in one pass.

Turns a messy keyword list into an optimized, deduplicated 250-byte Amazon backend search-term string with priority tiers.

Plan, code, render and stitch 3Blue1Brown-style math explainer videos with Manim Community Edition.

Drives the Superdesign CLI to design UI, slide decks, posters and image/video assets on an infinite canvas across multiple frontier models.

Traces where your recorded positions shifted over time, showing the old claim, the new claim, the dates, and the evidence that moved you.

Turns one focused Markdown behavior scenario into a deterministic test written in the repository's own test stack.

A stack-agnostic discipline skill that forces orient → baseline → smallest change → test → verify → self-review on every code change.

Send screenshots, UI mockups, or diagrams to external vision models (Doubao, Qwen, DeepSeek, OpenAI, Claude) and get a text description back.

Builds a full month of SMB social media posts with specific topics, angles, formats, and visual direction for each slot.

Stand up a 30–50 person Technical Advisory Board, interview them on repeat, and turn their exact words into positioning, roadmap and homepage copy.

A reference skill that teaches Claude the correct Python API for searching, renting, and managing GPUs on Vast.ai.

A skill that recommends chunk sizes, overlap ratios, and semantic boundary detection methods for RAG and vector search pipelines.

A reference skill for authoring Claude Code subagents: frontmatter rules, triggering examples, and system-prompt structure.

Turns "print me X" into a print-ready model for Bambu Lab printers: find or generate, repair, preview, hand off to Bambu Studio, then monitor.

Scans your codebase into a claude-db code graph and writes a five-note project profile covering stack, layout, conventions, workflows and architecture.

Build a small, real-world eval set to prove an agent or prompt works — and catch regressions whenever the model or prompt changes.

Audits a website for accessibility defects and guides fixes up to WCAG 2.1 AA compliance.

A second-pass editing skill that removes discourse-level AI tells — stated lessons, tidy single-track arcs, embodied emotion, vague allusions — that survive word-level humanizing.

A disciplined refactoring playbook that drives trace-mcp tools to assess risk, map impact, rename atomically, and verify results.

Generates 16 icons at once as a single 4×4 sheet, then slices each cell and removes the background to deliver a transparent-PNG icon set.