aesthetic-instrument
great_cto's own committed aesthetic — the instrument panel. Dark five-step surface ladder,…
Domain knowledge for SMB industry products — the vocabulary, the entities a spec must model, the incumbent to route around, and what a naive build gets wrong — one file per industry, plus local SEO and money-on-a-phone. Read the one file that applies before speccing or building
$ npx -y skills add avelikiy/great_cto --skill verticals --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/verticalsContext preview
The summary Claude sees to decide when to auto-load this skill.
Domain knowledge for SMB industry products — the vocabulary, the entities a spec must model, the incumbent to route around, and what a naive build gets wrong — one file per industry, plus local SEO and money-on-a-phone. Read the one file that applies before speccing or building
name: verticals description: "Domain knowledge for SMB industry products — the vocabulary, the entities a spec must model, the incumbent to route around, and what a naive build gets wrong — one file per industry, plus local SEO and money-on-a-phone. Read the one file that applies before speccing or building the product." when_to_use: | Speccing or building a product for one of the industries below, a public page that must rank locally, or an app that moves money on a phone. Read ONE file, not the directory. effort: low allowed-tools: Read, Glob
Each file below is the domain brief for one kind of product. They are references, not separate skills: one line in every session's skill list instead of twelve paragraphs, and the full text only when a product needs it.
| Product is for… | Read | |---|---| | HVAC, plumbing, cleaning, landscaping — dispatch, quoting, booking | `home-services.md` | | Agencies, consulting, creative — proposals, portals, time + invoicing | `professional-services.md` | | Restaurants, cafés — ordering, reservations, loyalty, shifts | `restaurants.md` | | SMB storefronts — catalog, inventory, pricing, cart recovery | `retail.md` | | Residential real estate — listings, lead CRM, transactions, property management | `real-estate.md` | | Studios, gyms, coaches — class booking, membership, content | `fitness.md` | | Creators, marketing — scheduling, analytics, monetization, sponsorships | `creator.md` | | HR, recruiting — ATS, onboarding, scheduling, surveys | `hr-recruiting.md` | | Contractors, field crews — bids, project logs, subcontractors | `construction.md` | | Shipping, warehouse-lite, routes, purchase orders | `logistics.md` | | An app that holds a balance or moves money on a phone (orthogonal to the rows above) | `fintech-mobile.md` | | Public pages that must be found in local or product search | `local-seo.md` |
From a great_cto checkout the files sit next to this one (`skills/verticals/<file>`). From an installed plugin, resolve the newest cached version:
# <file> is the row that applies, e.g. retail
cat "$(ls ~/.claude/plugins/cache/*/great_cto/*/skills/verticals/<file>.md 2>/dev/null | awk -F'/plugins/cache/' '{split($NF,p,"/"); print p[3], $0}' | sort -V | tail -1 | cut -d' ' -f2-)"Apply what it says to the spec: its vocabulary in the names, its must-model entities in the data model, its "what a naive build gets wrong" list as review items. If no row fits, read nothing — a near-miss vertical imports the wrong rules.
You already have the agent. This is everything around it. great_cto runs Claude Code as a pipeline of 70 specialist agents — an independent model checks each stage before the next builds on it, spending caps refuse rather than warn, and three decisions stay yours: what gets built, how, and whether it ships.
Repo: avelikiy/great_cto
great_cto's own committed aesthetic — the instrument panel. Dark five-step surface ladder,…
Catalogue of known SDLC anti-patterns that great_cto agents must actively reject when…
Analyze images, websites, and Figma files to extract their design and generate a `design.md`…
Shared review framework that every domain reviewer (pci, oracle, gov, edtech, healthcare,…
Structured idea generation + multi-LLM debate for the product-owner stage. Diverge (generate…
Run the great_cto controlled Codex lifecycle with controller-owned writes, verifier evidence,…