How it works

Three commands. The system does the rest.

No prompt-engineering. No agent-orchestration tutorial. No YAML.

01 · INSTALL

Drop into any repo

Detects archetype from manifests (25 archetypes · 15 detection signals), wires the gates, loads the right agents.

$ npx great-cto init ✓ archetype: web-service (95% confidence) ✓ security tier: standard ✓ 34 agents available ✓ ready in 12 seconds

02 · START

Describe the feature in one sentence

Architect drafts the architecture doc. You approve, refine, or reject. That's decision one.

$ /start "add 2FA via TOTP" → ARCH-2fa.md ready → pipeline: standard | LLM agent: ~45min · human team: ~2 days >> DECISION 1: approve? _

03 · REVIEW · QA · SECURITY

12 angles fire in parallel

Performance, security, SQL safety, concurrency, privacy, API contracts. Every finding rated. P0 blocks the gate.

✓ Performance · Security · SQL safety ✓ Concurrency · Privacy · API contracts ✓ Coverage 91% · 0 highs · CSO PASS >> DECISION 2: ship? _

04 · SHIP

Canary or done

5% → 20% → 100%. RELEASE doc auto-written. On-call notified. Memory updated for next time.

$ ship → canary 5% (3min) → 20% (5min) → 100% → RELEASE-2026-05-01.md → done.

The board you'll actually check

Six views. Real screenshots.
Live updates from your repo.

great-cto board at localhost:3141. Inbox · Kanban · Metrics · Agents · Memory · Public report. Vanilla HTML, zero deps — no Electron, no Tauri, no SaaS.

01 · KANBAN

Five columns. Uniform cards. Inline gate approval.

Gates · Backlog · In Progress · Done · Blocked. Cards are priority-coded, agent-tagged, with inline status / priority edit. ⌘K search across title / id / agent / labels. Live SSE — bd-CLI changes appear in the UI in <1s.

Filter bar: chip-toggle by agent, priority, label.
j / k navigation, Enter to open, ? for cheatsheet.
Click pipeline stage → drill-down to filtered kanban.

great_cto kanban board with 5 status columns

02 · METRICS

The numbers your CTO update needs.

Tasks shipped · LLM spend · cost-savings vs FTE · cycle time · QA pass rate · security blocks. 30-day daily-burn chart with budget alerts. A separate Agents tab shows per-agent time, LLM cost, and human-equivalent dollars at $150/hr.

Hero cards refresh every gate / verdict via SSE.
Cost panel pulls real plan data from docs/plans/.
Activity feed shows last 20 verdicts, cost-tagged.

great_cto metrics page with hero cards and cost chart

03 · AGENTS

Per-agent cost, time, and human equivalent.

34 specialist agents, each with its own time budget, LLM cost, and tasks-done counter. Compare to a human team at $150/hour and see the multiplier. Activity feed surfaces APPROVED / BLOCKED / FAIL verdicts with the agent that issued them.

agent cost breakdown with utilization bars and activity feed

04 · INBOX

Pick up where you left off.

Open a session — three columns greet you: In progress (your WIP tasks), Recent verdicts (what your agents finished while you slept), Decisions (every gate approval is logged with reasoning). Stop re-explaining your project to Claude.

Pending decisions — one-click approve / reject inline.
P0, blocked, stale (in-progress > 48h) auto-surfaced.
Append-only ~/.great_cto/decisions.md — query across all your projects.

great_cto inbox with Resume card and pending gates

05 · MEMORY

Four layers. Every Friday smarter.

PROJECT.md (archetype, goals) · lessons.md (per-project retros) · decisions.md (every gate approval with rationale) · verdicts/ (every agent verdict logged). Agents query memory before reading source files — solved problems stay solved.

06 · ALERTS

Email reaches you when the board can't.

Five things hit your inbox automatically: P0 incident, ship gate stale > 2h, security BLOCKED, budget at 80% / 100%, weekly Friday digest. Enter your email once, verify with a 6-digit code — no Resend signup, no API keys. Delivery routed through our relay, 100 emails/24h free per verified email.

🚨 P0 — payment-webhook 5xx spike (great_cto)

From: GreatCTO · 2 min ago

A P0 incident is open and needs your attention.

id demo-pmt-9k2
status in_progress
opened 2026-05-17 14:38

Claim P0 in board →

Why you ship without waking up at 2am

Twelve independent reviews.
Each finds what the other eleven miss.

Cursor and Copilot run one review pass. We run twelve. Every finding rated P0 / P1 / P2. P0 blocks the gate. You can't accidentally ship a SQL injection.

01 · PERFORMANCE

N+1, hot loops

"GET /orders runs 47 queries for 1 page — N+1 in OrderService.list()"

P1

02 · SECURITY

Injection, IDOR, JWT

"JWT verified but iss/aud claims not checked — token from any tenant valid"

P0

03 · READABILITY

Naming, complexity

"32-line nested ternary in checkout.ts:284 — extract to function"

P2

04 · SQL SAFETY

Raw interpolation

"Unparameterized ORDER BY in customers.ts:91 — injection vector"

P0

05 · LLM TRUST

Prompt injection

"User input flows into system prompt without sanitization (RAG step 4)"

P0

06 · SIDE EFFECTS

Mutation in conditions

"Mutation inside if() shadows logging — duplicate webhook events"

P1

07 · DATA PRIVACY

PII, GDPR/HIPAA

"Email logged in request middleware:54 — GDPR Article 5(1)(c)"

P1

08 · ERROR HANDLING

Swallowed exceptions

"try/catch around 3 unrelated failure modes — debugging blackhole"

P1

09 · CONCURRENCY

Races, deadlocks

"Cache stampede on /pricing — 1.4s tail at p99 under load"

P1

10 · DEPS

CVEs, abandonment

"lodash 4.17.21 has CVE-2026-1234 — bump to 4.17.22"

P1

11 · API CONTRACTS

Breaking changes

"PATCH /users now requires 'role' field — breaks v1 clients"

P0

12 · DESIGN SYSTEM

Tokens, a11y

"23 hex codes hardcoded — design tokens from src/theme.ts not used"

P2

vs the obvious alternatives

"Why not just use Cursor?"

Cursor and Copilot are great editors. They are not SDLC pipelines. Here's what each does — honestly.

	great_cto our pick	Cursor	Copilot Workspace	Claude Projects
SDLC orchestration
Multi-agent SDLC pipeline	✓ 34 specialists	✕	✕	✕
Auto archetype detection	✓ 25 archetypes	✕	✕	✕
12-angle code review	✓	⚠ single-pass	⚠ single-pass	✕
Compliance gates (PCI / HIPAA / SOX / EU AI Act)	✓	✕	✕	✕
Memory & visibility
Persistent memory	✓ decisions.md + verdicts	⚠ chat-only	✕	✓ chat scope
Multi-project view	✓	✕	✕	⚠
Public sharable reports	✓	✕	✕	✕
Ownership & cost
Open source	✓ MIT	✕	✕	✕
Runs locally	✓	⚠ partial	✕	✕
Pay your own API	✓	✕	✕	✕
Pricing	$0 + your API	$20/mo	$39/mo	$20/mo

We're not an editor — we orchestrate the process around your editor. Use Cursor inside the loop if you want.

Pricing

$34/month.
That's the whole bill.

GreatCTO is open source (MIT). You pay your own Anthropic API tokens. We don't see them. We don't bill you. Nothing to subscribe to.

Typical product team · 20 pipeline runs/month

quick · config / typo

$0.10

× 10

$1

quick · new endpoint

$1.00

× 6

$6

standard · feature

$5.00

× 3

$15

deep · cross-cutting

$12.00

× 1

$12

Total

~$34/mo

+ Routine triage auto-routed to Kimi K2 → 60–80% cost cut on log clustering
+ No per-seat. No SaaS. No data leaves your laptop.

Cursor Business

$40/seat/mo · ~$400/mo for a 10-eng team

Devin

$500/mo flat

GreatCTO

~$34/mo total · pay only when you ship

Quick start

30 seconds. No signup.
No credit card.

01 · init

$ npx great-cto init ✓ archetype detected ✓ 34 agents loaded ✓ ready in 12s

02 · wire into Claude Code

$ npx great-cto adapt \ --platform claude-code → CLAUDE.md → .great_cto/agents/ → MCP config ready

03 · start a feature

$ /start "add 2FA" → ARCH · review · QA · sec → DECISION 1? $ ship it → done.

CI gate

$ npx great-cto ci ./ \
   --sarif results.sarif

Auto-detects $GITHUB_ACTIONS — emits inline error annotations on PR diffs.

MCP server

$ npx great-cto mcp

Add to Claude Desktop config — exposes 5 tools to chat.

Webhook receiver

$ npx great-cto serve

HMAC-verified GitHub / Sentry receivers + Slack/Discord/PagerDuty fan-out.

FAQ

The questions everyone asks before installing.

Will it commit to my repo without me knowing?

No. Every commit goes through your local git. Two human gates per feature. You can audit every diff before approve.

What if it makes a mistake on the architecture?

You reject the gate. Architect re-drafts with your reasoning. The conversation is in your terminal — same as a normal Claude Code session.

How is this different from Cursor or Copilot?

Cursor and Copilot are editors — they write code on demand. GreatCTO sits one level above: it decides which agents to run, which gates to enforce, and which compliance checks to load. It runs inside Claude Code, orchestrating the full SDLC while you make two decisions per feature.

What about CI? Do I need a chat to use this?

No. npx great-cto ci ./ drops into GitHub Actions / GitLab / any CI as a single step. Auto-detects $GITHUB_ACTIONS and emits inline ::error file=... annotations on PR diffs. Outputs SARIF for the GitHub Security tab + JUnit XML for test reporters. Exit 0 clean, 1 findings, 2 setup error.

Does it work on existing codebases?

Yes. /audit reads the repo, builds CODEBASE.md, generates a backlog of gaps. Tested on JS/TS, Python, Rust, Go. ~2 minutes for 100k LOC.

What about my secrets?

Nothing leaves your machine except Claude API calls (your tokens, your Anthropic account). The board, the memory, the patterns — all local files in .great_cto/. You can .gitignore them or commit them — your call.

Can I disable an agent I don't need?

Edit .great_cto/PROJECT.md → agents: [...]. Or override at runtime: /start "feature" --agents=architect,senior-dev,qa.

I'm not solo — I have one cofounder / one contractor. Can we both use it?

Yes, in "thin team" mode. One git repo, both cd into it, both run claude. .great_cto/ and Beads tasks live in git. There's no shared dashboard, no auth, no per-seat licensing — and that's intentional. great_cto is built for the one-person engineering org. If you need shared boards, multi-seat auth, or per-developer cost attribution, you've outgrown it.

What happens if Anthropic changes pricing?

Routine triage already auto-routes to Kimi K2 (cheaper Sonnet-equivalent). You can pin a specific model in PROJECT.md. Worst case: it costs 2× for 6 weeks until we add the next provider. The plugin is MIT — you can fork.

Stop being the only person
who can ship.

Three commands. The system does the rest.

Drop into any repo

Describe the feature in one sentence

12 angles fire in parallel

Canary or done

Six views. Real screenshots.
Live updates from your repo.

Five columns. Uniform cards. Inline gate approval.

The numbers your CTO update needs.

Per-agent cost, time, and human equivalent.

Pick up where you left off.

Four layers. Every Friday smarter.

Email reaches you when the board can't.

Twelve independent reviews.
Each finds what the other eleven miss.

"Why not just use Cursor?"

$34/month.
That's the whole bill.

Typical product team · 20 pipeline runs/month

30 seconds. No signup.
No credit card.

The questions everyone asks before installing.

The bottleneck is you.
Stop being it.

Stop being the only person who can ship.

Three commands. The system does the rest.

Drop into any repo

Describe the feature in one sentence

12 angles fire in parallel

Canary or done

Six views. Real screenshots.Live updates from your repo.

Five columns. Uniform cards. Inline gate approval.

The numbers your CTO update needs.

Per-agent cost, time, and human equivalent.

Pick up where you left off.

Four layers. Every Friday smarter.

Email reaches you when the board can't.

Twelve independent reviews.Each finds what the other eleven miss.

"Why not just use Cursor?"

$34/month.That's the whole bill.

Typical product team · 20 pipeline runs/month

30 seconds. No signup.No credit card.

The questions everyone asks before installing.

The bottleneck is you.Stop being it.

Stop being the only person
who can ship.

Six views. Real screenshots.
Live updates from your repo.

Twelve independent reviews.
Each finds what the other eleven miss.

$34/month.
That's the whole bill.

30 seconds. No signup.
No credit card.

The bottleneck is you.
Stop being it.