Live on Canopy.
Think of it like this: your business has departments — research, marketing, sales, launch, scale. Each department has a head and a team. Each team follows your playbooks, your scripts, your standards.
Growth Operator Agency is that whole org chart, written down as a folder of plain markdown files.
Open the folder. Hand it to any AI agent — Claude, ChatGPT, Cursor, OpenClaw. The agent reads SYSTEM.md, learns who it is, learns your business, and starts working a department.
You stay in the loop on the calls that need your judgement. The rest runs while you sleep.
| Step | What happens |
|---|---|
| 01 | Clone the folder |
| 02 | Fill in company.yaml with your business — offer, ICP, voice, pricing |
| 03 | Hand the folder to any agent. It reads SYSTEM.md, becomes a growth operator, and starts shipping |
| Without | With |
|---|---|
| You write every VSL, every email, every ad — because no one else can match your voice. | The voice is encoded. Agents draft in your patterns. You approve. |
| The closer leaves at month 8. The pipeline collapses. | Sales judgement is in agents/sales/. The next closer onboards in a day. |
| The VA copies last quarter's launch and the offer falls flat. | plan-launch reads your offer, your audience, your win patterns — and ships a launch built on the work that actually converted. |
| Quality decays the moment you stop reviewing every asset. | Every output passes the blind output test before it ships. Below 0.8 S/N, it's rejected automatically. |
| Your business stops when you stop. | Your business runs while you sleep, with receipts. |
git clone https://github.com/Heuresis/Growth-Operator-Agency.git your-workspace
cd your-workspace
./boot.shBoot prints the workspace credentials and runtime install options. Then install into your runtime of choice:
./scripts/install.sh --tool claude-codeEleven runtimes supported. See Multi-tool integrations.
Fill in company.yaml with your business context. Then ask for what you need:
/research a market research brief
/build-icp a customer profile
/design-offer an offer document
/build-vsl a video sales letter
/build-funnel a 7-layer funnel
/ad-creative hundreds of ad variants
/plan-launch a launch plan
/build-sop a team SOP
Full setup walkthrough: Quickstart · 30 minutes.
Every Heuresis workspace is the same shape. A boot layer reads your business context, then activates the org-chart, skill outputs, methodology brain, and trigger manifest. Eleven runtime integrations ship — see Multi-tool integrations.
┌──────────────────────────────────────────────────────────────┐
│ THE ENCODED WORKSPACE │
│ │
│ ┌───────────┐ ┌───────────┐ ┌───────────┐ ┌───────────┐ │
│ │ SYSTEM.md │ │ ENCODING │ │ INVARIANTS│ │ company │ │
│ │ boot file │ │ schema │ │ rules │ │ context │ │
│ └───────────┘ └───────────┘ └───────────┘ └───────────┘ │
│ │
│ ┌───────────┐ ┌───────────┐ ┌───────────┐ ┌───────────┐ │
│ │ agents/ │ │ skills/ │ │ reference/│ │ triggers │ │
│ │ org chart │ │ outputs │ │ brain │ │ manifest │ │
│ └───────────┘ └───────────┘ └───────────┘ └───────────┘ │
└──────────────────────────────────────────────────────────────┘
│
▼
any agent runtime that reads files — 11 ship
growth-operator-agency/
│
├── README.md ← the pitch
├── SYSTEM.md ← boot file · any AI becomes a growth operator
├── INVARIANTS.md ← 28 sacred rules
├── ENCODING.md ← 11-compartment schema
├── company.yaml ← YOUR business · fill once
│
├── agents/ ← 41 specialists · an org chart of a real company
│ growth-ceo top orchestrator
│ 7 department heads · 33 specialists
│
├── skills/ ← 36 capabilities · each produces one asset
│ │
│ │ FOUNDATIONS
│ ├── /research Market Research Brief
│ ├── /build-icp Ideal Customer Profile
│ ├── /build-positioning Positioning Document
│ ├── /design-offer Offer Document
│ └── /extract-voice Brand Voice Architecture
│ │
│ │ MARKETING
│ ├── /content-calendar 30-day content plan
│ ├── /ad-creative Paid ad brief (8 ad types)
│ ├── /write-linkedin-post LinkedIn post
│ ├── /write-reel Short-form script (10 frameworks)
│ ├── /write-youtube Long-form video (7 types)
│ ├── /write-x-thread X thread (7 types)
│ └── /story-sequence IG story sequence
│ │
│ │ NURTURE
│ ├── /lead-magnet Lead magnet (9 types)
│ ├── /email-sequence Email sequence (8 types)
│ ├── /webinar-script Webinar / challenge (6 formats)
│ └── /post-booking-nurture Show-rate stack
│ │
│ │ SALES
│ ├── /build-vsl Video Sales Letter (5 variants)
│ ├── /build-funnel Funnel Architecture (15 archetypes)
│ ├── /sales-script Discovery → pitch → close
│ ├── /objection-library 20+ objections × reframes
│ ├── /call-prep Pre-call brief
│ ├── /proposal Post-call proposal
│ ├── /application-form Qualifying application
│ ├── /tripwire-design $7–$97 entry offer
│ ├── /landing-page 4-page stack
│ └── /competitor-intel Competitor teardown
│ │
│ │ LAUNCH · SCALE · PARTNERSHIPS
│ ├── /plan-launch · /launch-report · /build-sop · /hiring-brief
│ ├── /retention-check · /case-study · /revenue-report
│ └── /jv-webinar-proposal · /affiliate-program · /referral-program
│
└── reference/ ← the brain that makes skills smart
├── frameworks/ 99 methodology docs
├── operators/ 18 archetyped playbooks
├── platforms/ 6 platform playbooks
├── swipe-file/ 170+ hooks · templates · threads · testimonials
├── templates/ 20 output templates
├── verticals/ 11 vertical packs
└── legal-templates/ 11 starter legal docs
Each file is plain text. Each folder is owned by you. Nothing is locked behind an app.
Your creator business, in one folder.
- Landing page — multiple variants, live A/B tested
- Video sales letter — embedded in the landing page, scripted from your ICP and offer
- Email sequence — 7 emails, scheduled across LinkedIn, X, and email
- Retargeting ads — hundreds of creatives, tested by angle
- Sales call script — tuned on objection patterns captured from your calls
- Follow-up sequence — triggered 48 hours after every call
- Upsell page — shipped once the funnel proves out
Every action leaves a receipt. Every variant tracks its own conversion. Every winner gets promoted. Every paused variant tells you why.
Wire the workspace into a runtime with cron, webhook, and event triggers. The agents keep working without you in the room.
- Monday 09:00 (cron) — retention-agent runs a health check on every active client
- On
lead.captured(webhook) — ICP-match scoring fires, nurture sequence routed - On
launch.cart.closed(event) — launch-report generated within 48 hours - Daily 06:00 (cron) — content-agent queues 4 LinkedIn posts, 2 short-form hooks
- Continuous (event) — retargeting-agent swaps creatives based on CTR
Triggers are declared in paperclip.manifest.yaml. Wire them to your scheduler and the workspace operates as a 24/7 growth team.
7 agent-first departments. 41 agents. 36 skills.
- Foundations — research · ICP · niche · offer · brand voice
- Marketing — content · YouTube · short-form · X · LinkedIn · stories · paid ads · SEO · podcast
- Nurture — email sequences · lead magnets · community · webinars · SMS
- Sales — VSL · funnel · sales scripts · DM sales · call prep · proposal · CRM
- Launch — launch manager · post-launch analyst
- Scale — SOPs · team hiring · competitor intel · financial · retention · case studies
- Partnerships — JV webinars · affiliate · referral
Each department is a full department rebuilt as agent-first — agents that carry your judgement, read signals, and make decisions.
The workspace is files. Files run anywhere that reads files. The Agency ships conversion + install scripts so the same agents work across every major agentic coding tool — no vendor lock, no rewrite.
| Tool | Install | Destination |
|---|---|---|
| Claude Code | ./scripts/install.sh --tool claude-code |
~/.claude/agents/heuresis-growth-operator-agency/ |
| GitHub Copilot | ./scripts/install.sh --tool copilot |
~/.github/agents/ + ~/.copilot/agents/ |
| Antigravity (Gemini) | ./scripts/install.sh --tool antigravity |
~/.gemini/antigravity/skills/ |
| Gemini CLI | ./scripts/install.sh --tool gemini-cli |
~/.gemini/extensions/heuresis-growth-operator-agency/ |
| OpenCode | ./scripts/install.sh --tool opencode |
./.opencode/agents/ |
| Cursor | ./scripts/install.sh --tool cursor |
./.cursor/rules/ |
| Aider | ./scripts/install.sh --tool aider |
./CONVENTIONS.md |
| Windsurf | ./scripts/install.sh --tool windsurf |
./.windsurfrules |
| OpenClaw | ./scripts/install.sh --tool openclaw |
~/.openclaw/heuresis-growth-operator-agency/ |
| Qwen Code | ./scripts/install.sh --tool qwen |
./.qwen/agents/ |
| Kimi Code | ./scripts/install.sh --tool kimi |
~/.config/kimi/agents/ |
Five tools (Antigravity, Gemini CLI, OpenClaw, Cursor, Kimi) require format conversion — the install script handles it transparently. To inspect the converted artifacts before deploying:
./scripts/convert.sh --tool <tool-name>
ls build/<tool-name>/Every tool has its own README with activation patterns, customization notes, and uninstall steps. Full list: integrations/README.md.
Runtime-swappable. Your workspace is the asset. The runtime is replaceable.
What every Heuresis workspace ships with, regardless of which runtime you point at it.
| Property | What it does |
|---|---|
| Plain-text workspace | Every agent, skill, and framework is markdown + YAML. Read it, fork it, version it, diff it. |
| Runtime-portable | The same files run across every major agentic tool. Eleven integrations ship — see Multi-tool integrations. |
| Goal ancestry | Every skill carries the company's mission, ICP, offer, and brand voice as live context. Agents act on the why, not just the what. |
| Encoded judgement | Agents carry the operator's decisions, priors, and quality standards. The workspace replicates the operator's judgement, not their task list. |
| Receipt trail | Every cycle leaves an artefact — a brief, a draft, a score, a diagnostic. Auditable end-to-end. |
| Compounding library | Wins, swipe-file additions, and benchmark updates feed back into the workspace. Every cycle sharpens the next. |
| Trigger-ready | Cron, webhook, and event triggers declared in a manifest. Wire to any scheduler — daily kicks, post-publish audits, lead-routing — and the workspace operates while you sleep. |
| Owned outright | MIT-licensed. Fork it, host it, run it forever. Yours. |
Every founder-led creator business eventually hits the same wall: the founder IS every department.
The closer leaves at month 8 and the pipeline collapses. The VA copies last quarter's launch and the offer falls flat. The VSL is the only one that books calls because the founder is the only one who can write it — so the calendar bottlenecks on a single human.
Encoding changes the shape of the week. Your judgement — pricing logic, qualification instincts, voice, quality standards, the patterns you carry in your head — gets written into agents that run each department on your behalf. The founder stays. The judgement scales. The business compounds.
Every cycle, each department gets sharper. The gap between your firm and every competitor operating off memory widens.
This is one template in the library. More shipping, vertical by vertical. Every outcome claim we publish traces to a real deployment with a real operator. Thesis, method, and source go public on ideas. Receipts wait their turn.
- ✅ 41 agents · 7 departments · 36 skills shipped
- ✅ 99 framework docs · 18 operator playbooks · 11 vertical packs
- ✅ 170+ swipe-file specimens · 20 output templates · 11 legal templates
- ✅ Eleven runtime integrations: Claude Code · Copilot · Gemini · Cursor · Aider · Windsurf · OpenClaw · Qwen · Kimi · OpenCode · Antigravity
- ✅ Trigger manifest: cron · webhook · event
- ✅ Receipt trail across every skill
- ⚪ Voice-drift detection across long-form copy
- ⚪ Multi-operator org configurations
- ⚪ Per-skill benchmarks library
- ⚪ Workspace versioning · diff · rollback tooling
- ⚪ Vertical packs: B2B SaaS · agency-services · info-product · e-com
- ⚪ Heuresis Cloud — managed hosting for the workspace
| Folder | What it is |
|---|---|
.claude/commands/ |
Claude Code slash-command shims — one .md per skill, wires /skill-name to its runtime |
.github/ |
GitHub-native config: issue templates, PR template, contribution guide |
agents/ |
41 agent role definitions — CEO, 7 department heads, 33 specialists |
docs/ |
User documentation: quickstart, architecture, FAQ, glossary, file tree, skill authoring guide |
prompts/ |
Domain-indexed prompt library: research, sales, content, email, ads, analytics, voice |
reference/ |
Methodology library: frameworks, operator playbooks, swipe files, templates, vertical playbooks, legal templates |
skills/ |
36 skill packages — each with SKILL.md, variants, examples, evidence, runtime adapters |
spec/ |
Workspace specifications: banned vocabulary, context thresholds, runtime contracts |
workflows/ |
Operational workflows: client onboarding, division pipelines, automations, handoffs, ops cadences |
- Quickstart — setup in 30 minutes
- Architecture — how the folder is built
- Skill Authoring — write your own agents and skills
MIT. Free forever.
Built by Syed Hussain at Heuresis.