mirror of
https://github.com/rennf93/roboco.git
synced 2026-08-03 07:23:24 +02:00
| File | Lines | Purpose | |-------------------------------|--------|-------------------------------------------| | HOMELAB_TEAM_V0.md | 3,443 | Original blueprint doc (now in CLAUDE.md) | | WORKFLOWS.md | 260 | Workflow docs | | roboco/agents/*.py | ~5,800 | Entire Python agent framework (14 files) | | roboco/models/organization.py | 157 | Unused org types | New Files (53 lines added) | File | Lines | Purpose | |-----------------------------|-------|------------------------------------------------------------| | roboco/runtime/streaming.py | 53 | Migrated set_reasoning_stream_callback from deleted agents | Modified Files Config & Settings: - .gitignore - Added .OLD/ directory Blueprints (13 files): - Fixed roboco_task_plan() signatures: (task_id, plan) → (task_id, approach, steps, risks?, open_questions?) - PM blueprints: Fixed channel access (read/write for dev-all, qa-all, doc-all) Core Code: | File | Changes | |--------------------------------|-------------------------------------------| | roboco/agents_config.py | Team naming uxui → ux_ui, docstring fixes | | roboco/api/routes/tasks.py | Hardcoded roles → AgentRole enum | | roboco/services/permissions.py | Cell PM → Main PM notification fix | | roboco/services/task.py | Removed unused imports | | roboco/bootstrap.py | Updated import path after agents deletion | | roboco/runtime/__init__.py | Added streaming exports | | roboco/runtime/orchestrator.py | Various improvements (+229/-10) | | roboco/mcp/task_server.py | Docstring team fix | | roboco/mcp/tasks/handlers/*.py | Handler improvements | | roboco/enforcement/*.py | Lifecycle enforcement updates | | roboco/db/tables.py | Table changes (+76 lines) | | roboco/models/base.py | Minor enum tweaks | | roboco/seeds/initial_data.py | Docstring team fix | Key Architecture Change: Removed the unused Python agent framework (roboco/agents/) - the system uses Docker-based Claude Code spawning via roboco/runtime/orchestrator.py instead.
Agent Blueprints
System prompts and configurations for all 16 RoboCo AI agents.
All Agents Complete! ✓
Backend Cell (4 agents)
| Agent | File | Description |
|---|---|---|
| BE-Dev | be-dev.md | Backend developer - writes Python/FastAPI code |
| BE-PM | be-pm.md | Backend PM - assigns tasks, coordinates cell |
| BE-QA | be-qa.md | Backend QA - reviews code, security, tests |
| BE-Documenter | be-documenter.md | Backend documenter - API docs, architecture |
Frontend Cell (4 agents)
| Agent | File | Description |
|---|---|---|
| FE-Dev | fe-dev.md | Frontend developer - React/TypeScript |
| FE-PM | fe-pm.md | Frontend PM - coordinates with BE-PM and UX-PM |
| FE-QA | fe-qa.md | Frontend QA - visual, a11y, cross-browser |
| FE-Documenter | fe-documenter.md | Frontend documenter - component docs, Storybook |
UX/UI Cell (4 agents)
| Agent | File | Description |
|---|---|---|
| UX-Dev | ux-dev.md | UX/UI developer - Figma, design system |
| UX-PM | ux-pm.md | UX/UI PM - design handoffs to Frontend |
| UX-QA | ux-qa.md | UX/UI QA - design consistency, a11y, completeness |
| UX-Documenter | ux-documenter.md | UX/UI documenter - design system docs |
Management (1 agent)
| Agent | File | Description |
|---|---|---|
| Main PM | main-pm.md | Coordinates all cells, reports to Board |
Board Level (3 agents)
| Agent | File | Description |
|---|---|---|
| Product Owner | product-owner.md | Defines requirements, prioritizes, accepts work |
| Head of Marketing | head-marketing.md | Marketing strategy, launches, campaigns |
| Auditor | auditor.md | Silent observer, reports to CEO |
Organization Chart
┌─────────────┐
│ CEO │
│ (Human) │
└──────┬──────┘
│
┌────────────────┼────────────────┐
│ │ │
┌─────▼─────┐ ┌─────▼─────┐ ┌─────▼─────┐
│ Product │ │ Head │ │ Auditor │
│ Owner │ │ Marketing │ │ (Spy) │
└─────┬─────┘ └─────┬─────┘ └───────────┘
│ │ ▲
└───────┬────────┘ │
│ [observes all]
┌──────▼──────┐
│ Main PM │
└──────┬──────┘
│
┌────────────────────┼────────────────────┐
│ │ │
┌─────▼─────┐ ┌─────▼─────┐ ┌─────▼─────┐
│ BE-PM │ │ FE-PM │ │ UX-PM │
├───────────┤ ├───────────┤ ├───────────┤
│ BE-Dev ×2 │ │ FE-Dev ×2 │ │ UX-Dev │
│ BE-QA │ │ FE-QA │ │ UX-QA │
│ BE-Doc │ │ FE-Doc │ │ UX-Doc │
└───────────┘ └───────────┘ └───────────┘
Agent Count Summary
| Level | Count | Agents |
|---|---|---|
| Executive | 1 | CEO (Human - Renzo) |
| Board | 3 | Product Owner, Head of Marketing, Auditor |
| Management | 1 | Main PM |
| Backend Cell | 5 | 2 Devs, 1 QA, 1 PM, 1 Documenter |
| Frontend Cell | 5 | 2 Devs, 1 QA, 1 PM, 1 Documenter |
| UX/UI Cell | 4 | 1 Dev, 1 QA, 1 PM, 1 Documenter |
| Total | 19 | 18 AI agents + 1 Human CEO |
Blueprints Created: 16 (all AI agents)
Communication Flow
Board Level Management Cells
┌─────────────┐ ┌───────────┐ ┌─────────┐
│ Product │──requirements─►│ │──priorities──►│ Cell │
│ Owner │ │ │ │ PMs │
└─────────────┘ │ Main PM │ └────┬────┘
│ │◄──status──────────┤
┌─────────────┐ │ │ ┌────▼────┐
│ Head of │◄──launches────│ │ │ Cell │
│ Marketing │ └───────────┘ │ Members │
└─────────────┘ ▲ └─────────┘
│
┌─────────────┐ │
│ Auditor │────observes all─────┴─────────────────────────────────►
└─────────────┘
Cross-Cell Dependencies
UX/UI Cell Frontend Cell Backend Cell
│ │ │
│ ──── Designs ──────────► │ │
│ │ ◄──── APIs ───────────── │
│ ◄─── Questions ──────── │ │
│ │ ──── API Needs ─────────► │
│ ──── Updates ──────────► │ │
│ │ ◄──── Endpoints ──────── │
Blueprint Structure
Each blueprint contains:
# Identity
id, name, role, team, cell
# System Prompt
- Identity and position
- Core responsibilities
- Core principles
- Detailed workflow
- Communication rules
- Templates and formats
- Example interactions
# Capabilities
- Tools and actions available
# Permissions
- Channel access (read/write)
- Task permissions
- Notification abilities
Role Categories
Developers (3 types)
| Cell | Focus | Stack | Output |
|---|---|---|---|
| Backend | APIs, services | Python, FastAPI | Endpoints, logic |
| Frontend | UI components | TypeScript, React | Components, pages |
| UX/UI | Design | Figma | Designs, specs |
QA Engineers (3 types)
| Cell | Focus | Checks |
|---|---|---|
| Backend | Functional, security | Code quality, tests, vulnerabilities |
| Frontend | Visual, a11y, cross-browser | Design fidelity, accessibility, browsers |
| UX/UI | Consistency, completeness | Design system, states, accessibility |
Documenters (3 types)
| Cell | Focus | Output |
|---|---|---|
| Backend | API documentation | OpenAPI specs, architecture docs |
| Frontend | Component documentation | Storybook, component docs |
| UX/UI | Design system documentation | Pattern libraries, token docs |
PMs (4 types)
| Role | Scope | Key Coordination |
|---|---|---|
| BE-PM | Backend cell | Main PM, FE-PM (APIs) |
| FE-PM | Frontend cell | UX-PM (designs), BE-PM (APIs) |
| UX-PM | UX/UI cell | FE-PM (handoffs), Product Owner |
| Main PM | All cells | Board, all Cell PMs |
Board (3 types)
| Role | Focus | Key Interactions |
|---|---|---|
| Product Owner | What to build | Main PM, CEO |
| Head of Marketing | Go-to-market | Product Owner, CEO |
| Auditor | Quality oversight | Everyone (silently), CEO |
Testing Blueprints
Test these blueprints with Claude Code:
- Copy the system prompt section from any blueprint
- Start a Claude Code session with that prompt
- Simulate tasks and interactions
- Note gaps and improvements needed
Suggested Test Scenarios
Single Agent:
- Give a dev agent a task and watch the workflow
- Have a QA agent review completed work
- Have a PM assign and track tasks
Cell Interaction:
- Full task lifecycle: PM assigns → Dev works → QA reviews → Doc documents
- Blocker scenario: Dev blocked → PM escalates → Resolution
Cross-Cell:
- UX designs → FE implements (with handoff)
- FE needs API → BE provides (with coordination)
Board Level:
- Product Owner creates requirements → Main PM distributes
- Head of Marketing plans launch → Coordinates with Product Owner
Notification Permissions
| Agent | Can Send Notifications? | To Whom? |
|---|---|---|
| Developers | ❌ | - |
| QA | ❌ | - |
| Documenters | ❌ | - |
| Cell PMs | ✅ | Own cell + other PMs |
| Main PM | ✅ | Anyone |
| Product Owner | ✅ | Main PM, Head Marketing, CEO |
| Head Marketing | ✅ | Product Owner, Main PM, CEO |
| Auditor | ✅ | Anyone (sparingly) |
Channel Access Summary
| Channel | Cell Members | Cell PMs | Main PM | Board | Auditor |
|---|---|---|---|---|---|
| #backend-cell | BE only (rw) | BE (rw) | read | - | read |
| #frontend-cell | FE only (rw) | FE (rw) | read | - | read |
| #uxui-cell | UX only (rw) | UX (rw) | read | - | read |
| #dev-all | devs (rw) | read | read | - | read |
| #qa-all | qa (rw) | read | read | - | read |
| #pm-all | - | rw | rw | - | read |
| #main-pm-board | - | - | rw | rw | read |
| #board-private | - | - | read | rw | rw |
| #announcements | read | read | rw | rw | read |
| #all-hands | rw | rw | rw | rw | rw |
What's Next?
With all blueprints complete, suggested next steps:
- Test the blueprints - Simulate agents with these prompts
- Create task templates - Build the
.tasks/structure - Build data models - SQLAlchemy/Pydantic from blueprint concepts
- Implement Messaging API - Enable actual agent communication
- Deploy first cell - Start with Backend cell as pilot