wagnersza

Internal helper contract for calling the copilot-companion runtime from Claude Code

Internal guidance for composing Copilot prompts for coding, review, diagnosis, and research tasks across GPT-5.4, GPT-5.3-Codex, and Gemini 3.1 Pro inside the Copilot Claude Code plugin

Internal guidance for presenting Copilot helper output back to the user

Author or update a DESIGN.md file — the plain-text format (YAML design tokens + Markdown prose) that describes a product's visual design system to coding agents, per the google-labs-code/design.md spec. Use this skill whenever the user wants to create, write, draft, or refine a DESIGN.md / design.md; capture a brand or visual identity in a portable spec; turn a vibe, brand brief, mood, or reference into a reusable design system; define design tokens (colors, typography, spacing, radii, components) for an LLM to build against; or extract/reverse-engineer a design system from existing code, a screenshot, a website, or a running UI. Trigger this even when the user doesn't say "DESIGN.md" by name but is clearly trying to pin down a consistent look-and-feel so generated UIs stay on-brand.

Diagnose and rewrite a draft prompt so a frontier model is engaged as a senior partner rather than a junior task-taker. Use when the user asks to improve, sharpen, review, fix, level-up, or rewrite a prompt; says things like "make this prompt better", "review my prompt", "is this a good prompt"; pastes a prompt and asks for feedback; or invokes `/prompt-improver`. Applies to prompts for heavy knowledge work (research, analysis, synthesis, strategy docs, PR FAQs, code planning) — not to short one-shot lookups or to agentic-pipeline prompts where strict structure and evals are the goal.

Internal helper contract for calling the cma-companion runtime from Claude Code