Blueprint<br>Agent-assisted product initiatives · MIT · on npm
Run a product initiative,<br>end to end.
AI-assisted projects move fast — and rot fast. Blueprint is a working method<br>for running a product initiative with an AI agent — research through<br>deployment. The agent does the building; the method supplies the checkpoints<br>and the paper trail. What ships is researched, prototyped, fact-checked, and<br>documented.
$ npx @nino-chavez-labs/blueprint-cli init --name=my-initiativeCopy
Quickstart ↓ Watch the demo → What is this? ↓ npm ↗ github ↗
plain language<br>The 30-second version.<br>Just the strip. Everything below goes deeper.
What it is
A method for doing product work with an AI carrying the heavy lifting.<br>The AI researches, prototypes, and drafts the documents; Blueprint is<br>the set of checkpoints that keeps the work verifiable.
Why it exists
AI-built work is fast and convincing — and that's the problem.<br>Confident claims nobody verified, decisions nobody wrote down.<br>Blueprint forces the receipts: every claim traced, every decision<br>recorded, every stage checked before the next one starts.
Who it's for
Anyone shipping product work with an AI agent — solo or in a team.<br>Solo, it's quality control: fast doesn't mean sloppy. In a team,<br>it's the harder problem: everyone has their own way of keeping AI<br>honest, and none of them line up. Blueprint is one way the whole<br>team can share.
the thinking behind it —<br>The kitchen I'm looking for ↗
how it works<br>One pipeline. Seven stages.
Each stage produces artifacts that feed the next. The insight:<br>prototype and documents are built simultaneously —<br>the prototype tests the decision, the docs capture the rationale.
01 Research<br>competitive analysis, codebase, comparables
02 Design<br>define scope — what the prototype will and won't do
03 Prototype<br>HTML pages matching the real product
04 Fact-check<br>every claim validated against source
05 Document<br>strategy, feasibility, integration plans
06 Deploy<br>one portal — docs + prototype + chat
07 Iterate<br>stakeholder feedback, triaged
proof<br>This site is Blueprint output.
Every page here was generated from a single blueprint.yml<br>by the same pipeline you’d run. The CLI is live on npm; the portal deploys from one config.<br>Blueprint runs on itself — the methodology, applied to productizing the methodology.
60gated updates<br>10decision records<br>15automated reviewers<br>12projects running Blueprint<br>7CLI commands<br>1config → whole portal<br>Open the portal →
get started<br>Up and running in five steps.<br>Five steps. The choices you make are written out below — every command runs as printed.<br>1 Prerequisites<br>Node 18+ and Claude Code. That’s it — there’s no server to stand up and nothing to host.
2 Scaffold your initiative<br>Generates a portal you own, validated against your config. --name is the only required flag — everything else defaults and is echoed back. Two starting points:<br>New idea — nothing exists yet. That’s all greenfield (the default) means.<br>$ npx @nino-chavez-labs/blueprint-cli init --name=my-initiativeCopy
Existing app — run it from that repo. That’s brownfield.<br>$ npx @nino-chavez-labs/blueprint-cli init --name=my-app --variant=brownfield --target=.Copy
3 Configure<br>Set the three things that matter in blueprint.yml — the rest has sane defaults.<br>variant: brownfield # greenfield | midstream | brownfield<br>tier: 1 # deliverable sophistication 0–2<br>portal_pattern: A # A = platform portal · B = redesign review
4 Run the pipeline<br>In Claude Code, the stages are skills. Run them in order — each gate is enforced by a reviewer.<br>/blueprint-research → /blueprint-prototype → /blueprint-validate → /blueprint-docs → /blueprint-deployCopy
5 Ship & check<br>Push and your portal deploys. Verify conformance — doctor is honest about what it didn’t check.<br>$ npx @nino-chavez-labs/blueprint-cli doctorCopy
Want the whole thing — stage by stage, every reviewer gate, on a real initiative? →<br>Run your first initiative,<br>or start at Learn.
reference<br>The CLI<br>7 commands, all real. npx @nino-chavez-labs/blueprint-cli — zero install.
blueprint init<br>Scaffold a new initiative and its portal — your variant, tier, and pattern choices validated before anything is written.
blueprint review<br>Run a stage gate: an automated reviewer that checks a stage is actually done. --list shows built-in + your own.
blueprint cost<br>Per-stage effort/model configuration and telemetry — dial spend up or down.
blueprint fleet<br>One report across every project running Blueprint: which are current, which have fallen behind.
blueprint upgrade<br>Move a project to the latest methodology version — preview first, dry-run by default.
blueprint doctor<br>Health check: loads the config, every reviewer, and the portal — and says what it didn’t check.
blueprint hive<br>Set up the shared workspace a team of agents coordinates through. hive setup shows its plan first — nothing is provisioned until you pass --execute.
ecosystem<br>Stay current. Feed back upstream.<br>Updates flow down without breaking you; your fixes and gaps...