T3 Stack is a TypeScript natural fit for web service / api. GreatCTO auto-detects both — adds the web-service archetype overlay, wires web-service-specific gates, and runs 83 specialist agents around your existing T3 Stack workflow.
GreatCTO reads your package.json and detects t3-stack + web-service archetype from signals: imports, file structure, env vars, README hints.
Attaches the web-service archetype overlay: OWASP API Top-10, GDPR data-minimization, SLO + error-budget gates. Override if your specifics differ; the defaults are sensible for T3 Stack-style projects.
qa-engineer runs tsc --strict / eslint / vitest --coverage; security-officer checks for prototype pollution + XSS sinks; performance-engineer reviews bundle size + cold-start times.
Bugs you've hit before in other T3 Stack projects (connection-pool exhaustion, ORM N+1 queries, retry storms) — the agent's Step 0 includes the prior detection order. MTTR drops 94 % on second occurrence (methodology).
$ cd my-t3-stack-app && npx great-cto init ✓ scanning manifests… found package.json ✓ stack: t3-stack (TypeScript) ✓ archetype: web-service ✓ overlay: applied ✓ 83 agents ready $ /start "add authenticated REST endpoint" ▸ architect drafting ARCH-web-service.md… ▸ pm decomposing into beads tasks… ⚐ gate:plan — your approval needed
Approve → 3 senior-devs run in parallel worktrees → 5 reviewers fan out in parallel → gate:ship → deploy. One real run walked stage-by-stage: /proof.
This is the shape of what senior-dev drafts for "authenticated REST endpoint" — auth first, schema validation, and the audit line the web-service reviewer requires before gate:ship opens.
// src/server/api/routers/webService.ts — reviewed by 5 agents
export const webServiceRouter = createTRPCRouter({
create: protectedProcedure // security-officer: auth before handler
.input(webServiceSchema) // qa-engineer: zod schema enforced
.mutation(async ({ ctx, input }) => {
const result = await handle(input, ctx.session.user);
await ctx.audit.log(ctx.session.user.id, 'authenticated REST endpoint',
result.confidence); // gate:web-service: every decision logged
return result;
}),
});web-service overlay.Public REST / GraphQL APIs with SLO gates and error budgets.
Internal platform APIs with authenticated service-to-service calls.
Partner-facing APIs with rate limiting and usage metering.
No black-box "AI does it all" loop. GreatCTO is a deterministic state machine — 8 stages, 22 nodes, 2 human gates. Every node maps to a real agent on GitHub. Inspect the state machine →
$ npx great-cto init
Free, MIT, runs locally. Built as a Claude Code plugin — install with one command.
Eight stages, two human gates, four memory layers. Why this exact shape, and what I tried that didn't work.
One run, one feature, from prompt to merged PR. Time, cost, and gate-by-gate breakdown — no marketing math.
Regex vs LLM-based archetype detection, the false-positive count, and why I keep rejecting the obvious fix.
The bottleneck in agentic SDLC isn't model quality — it's process governance. Here's the state machine that closes the gap.