rebar

When the user wants to create, adopt, extract, or lint a DESIGN.md for a project — a plain-markdown design system that AI coding agents read to keep UI visually consistent. Trigger on: 'design system,' 'DESIGN.md,' 'visual consistency,' 'brand the UI,' 'styling guide,' 'agent keeps changing the colors,' 'adopt a brand look.' Sits next to CLAUDE.md / AGENTS.md in a project root and prevents the three failure modes of agent-built UI: bootstrap default, color roulette, and style drift.

Exploratory web QA — drive a target URL like a real user, surface bugs / UX issues / regressions, write a structured report. Use when the operator runs `/dogfood {url}` or asks for a "QA pass," "exploratory test," or "smoke test the UI."

Makes AI-generated content sound genuinely human — not just cleaned up, but alive. Use when content feels robotic, uses too many AI clichés, lacks personality, or reads like it was written by committee. Triggers: 'this sounds like AI', 'make it more human', 'add personality', 'it feels generic', 'sounds robotic', 'fix AI writing', 'inject our voice'. NOT for initial content creation (use content-production). NOT for SEO optimization (use content-production Mode 3).

Full content production pipeline — takes a topic from blank page to published-ready piece. Use when you need to execute content: write a blog post, article, or guide end-to-end. Triggers: 'write a post about', 'draft an article', 'create content for', 'help me write', 'I need a blog post'. NOT for content strategy or calendar planning (use content-strategy). NOT for repurposing existing content (use content-repurposing). NOT for social captions only.

When the user wants to plan a content strategy, decide what content to create, or figure out what topics to cover. Also use when the user mentions "content strategy," "what should I write about," "content ideas," "blog strategy," "topic clusters," or "content planning." For writing individual pieces, see copywriting. For SEO-specific audits, see seo-audit.

When the user wants to write, rewrite, or improve marketing copy for any page — including homepage, landing pages, pricing pages, feature pages, about pages, or product pages. Also use when the user says "write copy for," "improve this copy," "rewrite this page," "marketing copy," "headline help," or "CTA copy." For email copy, see email-sequence. For popup copy, see popup-cro.

Extract text, tables, and images from PDFs (and scanned PDFs via OCR) using pymupdf for light/fast jobs and marker-pdf for OCR/equations/complex layouts. Use when an operator needs to pull content out of a client deck, contract, scanned form, research paper, or any PDF artifact — pick pymupdf for text-based PDFs (~25 MB install, instant) and marker-pdf only when OCR, equations, or layout reconstruction is required (3–5 GB models).

Gmail, Calendar, Drive, Docs, Sheets, and Contacts via per-client OAuth. Adapted from the Hermes Agent google-workspace skill. Use when an operator needs to read mail, schedule events, search Drive, read/write Sheets, or pull Docs content for a specific client. Auth is isolated per client — every client carries its own token under clients/{client}/.gws-token.json.

Geocode, find nearby POIs, compute travel distance/time, and get turn-by-turn directions using OpenStreetMap data (Nominatim, Overpass, OSRM) with zero API keys and Python stdlib only. Use when an operator or client task needs maps/location data without paying Google Maps fees — Velocity Electric routing, field-tech directions, "what's near this address," travel-time estimates, timezone lookups, or POI scouting for a region.

Edit PDFs in place using natural-language instructions via the `nano-pdf` CLI — fix typos, swap titles, update dates, rename clients on a specific page without rebuilding the document. Use when an operator needs a small, targeted text fix on an existing PDF (decks, contracts, reports) and rebuilding the source isn't worth it.

Optimize content to get cited by AI search engines — ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini, Copilot. Use when you want your content to appear in AI-generated answers, not just ranked in blue links. Triggers: 'optimize for AI search', 'get cited by ChatGPT', 'AI Overviews', 'Perplexity citations', 'AI SEO', 'generative search', 'LLM visibility', 'GEO' (generative engine optimization). NOT for traditional SEO ranking (use seo-audit). NOT for content creation (use content-production).

When the user wants to plan a product launch, feature announcement, or release strategy. Also use when the user mentions 'launch,' 'Product Hunt,' 'feature release,' 'announcement,' 'go-to-market,' 'beta launch,' 'early access,' 'waitlist,' 'product update,' 'GTM plan,' 'launch checklist,' or 'launch momentum.' This skill covers phased launches, channel strategy, and ongoing launch momentum.

Post, search, and manage X (Twitter) via the official xurl CLI. Use when you need to publish a tweet, attach media, reply, quote, search, like/repost/bookmark, follow, or send DMs from a script. Triggers: 'post to X', 'tweet from CLI', 'X API', 'xurl', 'cross-post to Twitter', 'post tweet with media'. NOT for LinkedIn/Facebook/Substack publishing — those use CDP helpers in spotcircuit-site/scripts/. NOT for analytics/scraping.

Debug Node.js (and tsx/TypeScript) interactively via `node inspect` REPL or scripted Chrome DevTools Protocol (CDP) clients. Use when console.log is insufficient — set real breakpoints, walk the call stack, dump locals/closures, evaluate expressions in paused frames, attach to a running process, or capture CPU/heap profiles. Pairs with rebar's CDP-based browser-harness.sh and complements superpowers:systematic-debugging (which is language-agnostic process; this is the Node-specific tooling reference).

Debug Python interactively via pdb (`breakpoint()`, `python -m pdb`, post-mortem) or remote-attach via debugpy/remote-pdb. Use when print/logging is insufficient — set breakpoints, step, walk the stack, dump locals, evaluate expressions, attach to a running daemon, or post-mortem an exception. Complements superpowers:systematic-debugging (which is the general process; this is the Python-specific tooling reference).