Your projects
DNA on your Mac
Pick a projects folder. Skeletor lists nested project folders, installs DNA, and shows which projects have DNA, Lab, or need an update.
Skeletor 1.0.0 · major release · out now
The control plane for every DNA project on your Mac.
One app. The whole fleet. Install DNA, keep Lab off until you want it, and send a project's memory to someone else — the same brain, ready to import.
- ✓Every project on your Mac, A to Z
- ✓DNA in. Lab stays out until you ask
- ✓Memory you can hand to a colleague
- ✓Exports land in Downloads
- ✓Hide any project. Uninstall DNA from one
- ✓Free forever on macOS
Open source · Built by Humaan
Git remembers your code. DNA remembers your system.
One dna doctor installs everything — 133 stem packs, rules, detection, CellularMemory, runtime, CI, and 1045 knowledge packs.Skeletor 1.0.0 installs DNA on your Mac and can hand a project's memory to a colleague.
“No amount of money ever bought a second of time.”
- 1.0
- Major release
- 1045
- Knowledge packs
- 133
- Prompt stem packs
- 1
- `doctor` install
Skeletor
The Mac control plane for your DNA projects.
Skeletor installs DNA on the projects on your Mac, keeps Lab off unless you add it to one project, and exports memory so a colleague can import the same brain.
Hand it to a colleague
Memory transplant
Export one .tar.gz of CellularMemory, knowledge, plans, workflows, Impressions, and feature history. Import replaces those folders on another machine.
- 01
Point at your fleet
Native folder picker. One projects root — every app DNA should know about.
- 02
DNA stays on the project
Missing .DNA/? Skeletor copies the kit. Lab and the runtime observer stay out until you add Lab to that one project.
- 03
Hand memory across
Export the project brain. A colleague imports the archive and those folders match yours.
Projects-folder first
Choose one root. Skeletor maps nested project folders and keeps each real project whole.
DNA without Lab by default
New installs copy the DNA kit and leave Lab and the runtime observer out. Add Lab to one project from the DNA page.
Memory transplant
Export a .tar.gz of memory, knowledge, plans, workflows, Impressions, and feature history. Import replaces those folders.
Marketplace
Knowledge packs and workflow bundles
Prefer the CLI? Install knowledge packs and workflow bundles from the marketplace — packs + stems + AI rules Cursor and Claude already understand.
dna marketplace install combo/nextjs-fullstack
combo/product-canvas-creation
Product canvas creation
Grounded product canvas from real architecture — not stub Impressions.
9 packs · 6 stems
combo/product-diagnosis-upgrade
Product diagnosis → competitor → upgrade leverage
Strategy ladder grounded with DNA CLI + CellularMemory + Impression Guard.
9 packs · 7 stems
combo/pmf-check
Product-market fit check
Validate PMF with research methods and lifecycle gates.
7 packs · 6 stems
combo/discovery-to-delivery
Discovery → engineering handoff
Research synthesis into tickets/specs for the Feature Factory.
8 packs · 7 stems
combo/shape-up-cycle
Shape Up betting cycle
Appetite, pitch, and delivery alignment for a six-week cycle.
6 packs · 6 stems
combo/nextjs-fullstack
Next.js fullstack app stack
Next.js + React + auth + DB + UI + observability patterns.
10 packs · 7 stems
The problem
Your repo knows the code. Your team knows the system. AI knows neither.
Premium AI memory still forgets
CellularMemory, Behaviour rules, and `.DNA/` live in git — cross-product context that survives every chat session and every new contractor.
Enterprise tools priced per seat and per service
One portfolio install: runtime observer, classified issues, 133 stems, quality gates, and knowledge packs — without Sentry × N and GitHub Teams on top of rent.
Five products, five amnesiac repos
Install on the squad folder once. DNA maps every app, shared auth, and dependencies — feature #5 uses the same template as feature #1.
Agents drift into feature branches and finish dirty
Agent Mesh is the control plane: a live registry, path claims, and Git Guardian hooks that physically block branch hops, worktrees, git add ., and uncommitted finishes.
What you get
One install. Stems, rules, detection, memory, runtime — nothing left out.
DNA scaffolds a machine-readable project layer your tools can trust — not another wiki that rots in Notion.
Purpose bundles
One command — `dna marketplace install combo/…` — installs linked knowledge packs, prompt stems, and AI rules for a whole workflow (Next.js, GDPR, PMF, billing…).
Project intelligence
`.DNA/` — neuralNetwork routing, 8 behaviour files, 7 CellularMemory regions, immune system. AI reads structure, not stale READMEs.
133 stem packs + rules
Guidelines, expectations, workflows in `.DNA/stems/` — plus AGENTS.md, Cursor + Claude skills, and `/dna-*` slash commands. Always on.
Automatic detection
Stack, monorepo apps, AI tools, GitHub, healthcare/legal domain, Impressions drift, auth patterns — doctor detects; you do not configure blind.
1045 knowledge packs
Curated packs — HIPAA, GDPR, frameworks, healthcare. Foundation installs from scan; marketplace adds depth; bundles group them by purpose.
DNA Lab
Production observability at /labs — Refresh disables + spins, Copy issue, analytics Overview. After every upgrade: dna update (or lab installs --fix), restart the API, hard-refresh /labs.
Runtime observer
Auto-wired into Express, Fastify, Next.js. Classified errors → contextual issues and repair plans. What broke, why, what to fix.
Feature factory
9-role agent loop, quality gates, CI + pre-push hook, Docker scaffold. Same delivery process for every feature — by policy, not luck.
Agent Mesh
Cursor agents register in SQLite, see each other's claims, and cannot create feature branches or finish dirty. `dna commit` stages only this agent's files.
Portfolio install
One squad. Many products. One install you never regret.
Most teams cannot afford Sentry on every service, premium AI memory on every seat, and GitHub Teams on top of rent. Install DNA once on the parent folder that holds every related app — not inside each repo.
“No amount of money ever bought a second of time.” — Howard Stark, Avengers: Endgame
# parent of product-a, product-b, shared-api…
cd ~/work/your-squad-folder
npx @superhumaan/dna-by-humaan doctor
Portfolio brain
One `.DNA/` at the squad folder — every related app, shared auth, API contracts, and cross-product maps.
133 stem packs
Guidelines, expectations, examples, and workflows in `.DNA/stems/` — plus `/dna-*` slash commands in Cursor and Claude.
Rules & skills
AGENTS.md, Cursor + Claude rules, Agent Mesh hooks, delivery pipeline, product-process, and 9-role feature factory — always on.
Auto detection
Stack, monorepo apps, AI tools, GitHub, healthcare/legal domain, Impressions drift, auth patterns, runtime entrypoints.
CellularMemory
Decisions, failures, patterns, system maps — persists in git. The memory subscriptions cannot export.
Runtime + immune system
Auto-wired observer, classified issues, repair plans — what broke, why, what to fix. Never auto-merge.
Delivery pipeline
CI workflows, pre-push quality hook, Docker scaffold, GitHub browser login — ship with gates, not guesswork.
Skeletor bridge
When Skeletor is installed, DNA pulls the local fleet by default and feeds sibling labs into AI context — cross-coordinate without re-explaining.
1045 knowledge packs
Foundation from scan + marketplace — HIPAA, GDPR, frameworks, healthcare country overviews, offline fallback.
Full manifest: Portfolio install guide · 133 stem packs · Marketplace
AI-native by design
Stop re-explaining your project every session
Run dna context cursor (or copilot, claude, windsurf) and your AI tool gets behaviour rules, stack constraints, open incidents, and installed knowledge packs — not a stale README.
- Cursor & Copilot
- Claude Code
- HIPAA & GDPR packs
- Portfolio install
Quick start
Up and running in minutes
Install from npm — no cloning required. Ask Cursor, a Cloud Agent, or Claude to set DNA up for you, or run three CLI commands in any TypeScript project. Then add a purpose bundle when you know the workflow.
Multi-product squad? Run doctor on the parent folder (e.g. all enterprise health apps under one directory) — not inside each repo. One brain remembers every product and how they connect.
Easiest — ask Cursor, Cloud, or Claude
No cloning, no CLI memorization. Paste a prompt and your agent runs dna doctor — the full platform install.
Cursor
Paste in Cursor Agent chat
“Install DNA by Humaan on this portfolio folder from npm — no cloning. Run: npx @superhumaan/dna-by-humaan doctor from the parent folder that holds all related apps (not inside each repo). Doctor scaffolds everything: 133 stem packs, Cursor + Claude rules, detection, CellularMemory, runtime, CI, and feature factory. Then run analyze and context cursor. Summarize what was installed and what DNA detected across the suite.”
Cursor Cloud
Paste in a Cloud Agent task
“Set up DNA by Humaan on this repo from npm only (package: @superhumaan/dna-by-humaan). Run npx @superhumaan/dna-by-humaan doctor at the project root (or portfolio parent if this is a multi-app squad folder). Do not run init separately unless doctor fails. After doctor: analyze, scan, and context cursor. Summarize stems, rules, detection results, and recommended marketplace packs. Do not clone the DNA monorepo.”
Claude
Paste in Claude Code or Claude Agent
“Install DNA by Humaan from npm — no cloning. Run npx @superhumaan/dna-by-humaan doctor at the folder root. One doctor installs: 133 stem packs, AGENTS.md, Cursor + Claude rules/skills, stack detection, runtime auto-wire, CI, feature factory, and foundation knowledge packs. Use real shell commands only. Then analyze and tell me what DNA found across all apps.”
Or install from npm / CLI
One command in any TypeScript project — or at the portfolio parent for squads with many related apps. Everything installs together; browse the marketplace later for domain depth.
01
npx @superhumaan/dna-by-humaan doctorOne command — `.DNA/`, 133 stems, rules, detection, runtime, CI, feature factory, foundation packs02
npm installIf doctor added @superhumaan/dna-by-humaan to package.json03
npx @superhumaan/dna-by-humaan analyzeMap every app under the folder — auth, shared libs, IVF gaps04
npx @superhumaan/dna-by-humaan context cursorFull portfolio context — open Cursor at any child appUpgrade DNA Lab (every release)
/labs is served by the package your API process resolves. Nested installs and long-lived Node processes leave Lab on a stale UI unless you finish these steps.
npx @superhumaan/dna-by-humaan@latest updateUpgrade CLI, packs, and nested Lab packagesnpx dna lab installs --fixOr force @latest in every owner package.json dir- Restart every Node process that mounts Lab (API or
dna lab serve) - Hard-refresh
/labs(Cmd+Shift+R) - Confirm
curl -s <api>/api/dna/labs/health | jq '{dnaVersion,labUi}'
DNA is free and open source. Sponsorship keeps the marketplace and infrastructure running.
Sponsor DNASkeletor 1.0.0
Download Skeletor for DNA on your Mac, and a memory archive a colleague can import.