orchestrationgencreator-stacklisted

Catalogs, scaffolds, and audits the per-repo tech stack across every brand (Arcanea, FrankX, Business, Starlight, verticals). Four verbs — audit (walk repos + emit constellation), scaffold (drop STACK.md template into a target repo), assign (bind tools to substrate/reasoning/coding/research/ops tiers), diff (surface drift between STACK.md intent and reality from .mcp.json + .claude/ + settings.json + installed CLIs). Use this skill whenever the user mentions stack architecture, repo inventory, "which AI assistant for which repo", custom GPTs, MCP routing across brands, browser workspaces per project, or asks "do we have a clean view of our tech stack". Default skill for the Starlight Architect when the question spans repos rather than a single one.
frankxai/Starlight-Intelligence-System · ★ 5 · AI & Automation · score 83

Install: claude install-skill frankxai/Starlight-Intelligence-System

# gencreator-stack Skill — Cross-Brand Stack Catalog + Drift Sentinel > *Every repo is an agent surface. Every brand is a constellation of surfaces. This skill is how you keep the map honest.* ## Activation Auto-activates when the user's question spans more than one repo or asks about tool assignment across the ecosystem. Pair with `starlight-architect` (Leadership tier). For single-repo decisions, defer to `engineering:architecture` (single ADR). For new brand spawning, compose with `integration/domain-stack-architecture` (vertical-tier scaffolder). ## Why this exists Frank operates 6+ active repos across 4+ brands. Each surface has a coding agent (Claude Code / Codex / Cursor / Gemini CLI), a reasoning surface (Claude Project / ChatGPT Project / Gemini Gem / Custom GPT), a research browser (Arc Space / Chrome+CIC / Comet), and a stack of native MCPs (Vercel / Supabase / Notion / Linear / Figma / GitHub / Gmail / Calendar). Without a manifest, three failure modes compound: 1. **Surface drift** — a repo's `CLAUDE.md` claims a tool that's no longer wired. Claude Code in that repo runs with stale context. 2. **MCP overlap waste** — same connector configured twice in two repos with conflicting auth scopes. Quiet rate-limit failures. 3. **Cognitive overhead** — Frank has to remember which AI runs which brand. Cognitive load grows non-linearly with brand count. The skill makes the manifest first-class: one `STACK.md` per repo, rolled up into one constellation file at substr