Short answer: both. Cursor in-editor multi-file code generation with deep Anthropic + OpenAI integration. GreatCTO orchestrates the SDLC around it — gates, parallel reviewers, archetype-specific compliance. Same plugin works inside Cursor.
regulated archetype → wires the right compliance gatesRun npx great-cto init in your regulated project. GreatCTO scans manifests, picks the archetype, attaches the right reviewer agents and compliance gates. You don't write the gates; you override them if your specifics differ.
Drafts ARCH.md + ADR + cost estimate. You approve scope at gate:plan. No implementation starts before your approval.
Cursor does the editing. GreatCTO orchestrates which agents claim which tasks (from the PM decomposition), runs them in isolated worktrees, and feeds the diff to reviewers.
qa-engineer · security-officer · performance-engineer · regulated-reviewer · code-reviewer. Verdicts aggregate to a single APPROVED / BLOCKED chip at gate:ship.
P0 incidents extract a lesson. Pattern hash + detection order written to .great_cto/lessons.md. Next iteration's agents read this in Step 0.
Full state machine with every node clickable to its agent on GitHub: /architecture.
Every box on the diagram is a clickable link to the agent's source on GitHub.
A real pipeline run walked stage by stage — timeline, LLM cost, e2e assertions, public artifacts.
47 paired P0 incidents · 4 memory-miss cases documented · raw data under NDA.
$ npx great-cto init ✓ scanning manifests… ✓ archetype: regulated ✓ adapting for: Cursor ✓ 83 agents ready
Free, MIT, runs locally. You pay your own LLM API. No SaaS dashboard, no telemetry by default.
Eight stages, two human gates, four memory layers. Why this exact shape, and what I tried that didn't work.
The bottleneck in agentic SDLC isn't model quality — it's process governance. Here's the state machine that closes the gap.
One run, one feature, from prompt to merged PR. Time, cost, and gate-by-gate breakdown — no marketing math.
47 paired P0 incidents across 12 repositories. 4 honest misses. Full methodology + how to replicate the measurement in your own repo.