Structure creates quality.
Discovery → planning → execution as distinct phases with review gates between them. Code without a plan is lumber without blueprints.
Synthex is fifteen specialised AI agents and eleven commands that take software from a messy idea to a peer-reviewed plan to production-ready, parallel-executed code — without you writing a for loop.
/plugin marketplace add bluminal/lumenaiThree convictions that the entire system is built around — they cascade into every command, agent, and quality gate.
Discovery → planning → execution as distinct phases with review gates between them. Code without a plan is lumber without blueprints.
One agent asked to "build a login page" produces something that works. Fifteen specialised roles produce something that's production-ready.
Git worktrees isolate every task. Three features get implemented at once without conflict. A sprint becomes an afternoon.
Every task maps to a requirement, every change maps to a task, every architectural decision is recorded. Five phases, one continuous loop.
Brain-dump → conversation → PRD that captures the why.
Milestones, phases, tasks. Peer-reviewed before a line of code.
Parallel worktrees, specialist agents, mandatory quality gates.
Performance audited, observable, and operationally ready.
DORA, retros, and follow-through loops back into discovery.
Modelled on the roles a real software startup would hire — clear scope, structured delegation, honest peer review.
Full-stack orchestrator. Breaks tasks down, delegates to specialists, rolls up the result.
Coordinates frontend delivery, enforces UX quality, accessibility, and design-system fidelity.
Interactive requirements gathering. Transforms PRDs into milestone-based plans.
System design, ADRs, plan-review feasibility, technology trade-offs.
Craftsmanship review — correctness, maintainability, reuse, conventions.
Mandatory gate. Secrets, injection, auth, supply chain, OWASP.
Test strategy, gap analysis, behavior-driven test authoring.
Plan review with cost, blast-radius and security verdict for AWS.
Token ownership, component governance, compliance audit.
Quantified, full-stack performance analysis with budgets.
SLOs, observability, runbooks, deployment risk, postmortems.
API docs, guides, migrations, changelogs verifiable against code.
Personas, journey maps, opportunity-solution trees, heuristic eval.
DORA + HEART + AARRR + OKRs. Trend, benchmark, learn.
Planned-vs-actual, blameless retros, tracked improvement loops.
Each command orchestrates a subset of agents. Mix them, chain them, or let next-priority run the whole loop.
"I'm still in control when I need to be. I'm still monitoring outputs, validating decisions. Claude provides the autopilot — but I'm the pilot. I decide when to take the stick."
Synthex is markdown all the way down. No runtime, no SDK, no API keys to manage — your existing Claude Code session is the engine.
Inside Claude Code, register the LumenAI marketplace where Synthex lives.
/plugin marketplace add bluminal/lumenaiScaffolds .synthex/config.yaml and the standard docs/ tree.
/synthex:initWrite a PRD with the Product Manager, generate a peer-reviewed plan, then let parallel Tech Lead instances execute.
/synthex:next-priority