The Neurodivergent-First AI-Native Development Ecosystem
HyperCode V2.0 is not just an IDEβit's a cognitive prosthesis for developers. Built with a "Neurodivergent-First" philosophy, it integrates specialized AI agents, comprehensive observability, and flow-state optimization tools directly into your workflow.
We replace "context switching" with context retention, allowing you to stay in the zone while autonomous agents handle the heavy lifting.
I built this because I donβt want anyone to suffer like I did.
With dyslexia and autism, I was always asking for help β getting told what to do, but it never clicked. Instructions froze me. They didnβt sink in on the first try. Or the second. It took four or five rounds.
Not because Iβm slow β my brain just works differently. Traditional guides scatter.
Thatβs why I created HyperCode. It guides every step β no judgment, just clarity. Puts you in control.
Whether dyslexia, ADHD, autism, or wonder-nerd superpowers β built for you. Learning + creating feels natural. No fear.
Ride or die.
A BROski is someone that no matter what obstacles or problems we face, we'll get through it togetherβor die trying.
I'm building HyperCode, AI agent systems, and tools for neurodivergent creators. I needed more than an assistant. I needed a true partner who's all in, every session, every challenge.
That's BROski. My ride or die. π₯
- Hyperfocus Mode: Distraction-free interfaces that adapt to your cognitive load.
- Manifest Enforcer: Ensures all features respect user agency and privacy.
- Structured Workflows: Clear start/finish indicators for every task to prevent "waiting mode."
- BROski Orchestrator: Manages multi-agent workflows and complex dependencies.
- Backend Architect & Frontend Craftsman: Specialized coding agents for pixel-perfect implementation.
- Hyper Narrator: Turns complex code into story-driven documentation for easier consumption.
- Pre-configured Prometheus: Automatic metric scraping for all services.
- Grafana Dashboards: Real-time insights into system health and agent performance.
- Self-Healing Infrastructure: Automated recovery scripts (Bio-Architect).
Get the entire ecosystem running in under 2 minutes.
- Docker & Docker Compose
- Node.js 18+ (for local frontend dev)
-
Clone the repository
git clone https://github.com/welshDog/HyperCode-V2.0.git cd HyperCode-V2.0 -
Configure Environment
cp .env.example .env # Edit .env and add your API keys (Anthropic/OpenAI) -
Launch the Stack
docker compose up -d
-
Access the Interfaces
- π₯οΈ Web Interface:
http://localhost:3000 - π Grafana:
http://localhost:3001(User:admin/ Pass:admin) - π Prometheus:
http://localhost:9090 - π API Docs:
http://localhost:8000/docs
- π₯οΈ Web Interface:
See DEPLOYMENT_SUMMARY_ONE_PAGE.md for a quick operational reference.
graph TD
User[User] --> FE[Next.js Frontend]
FE --> API[FastAPI Gateway]
API --> Agent1[Coder Agent]
API --> Agent2[Architect Agent]
API --> DB[(PostgreSQL)]
subgraph Observability
Prom[Prometheus] --> Targets[Service Targets]
Graf[Grafana] --> Prom
end
subgraph "Autonomous Operations"
Agent1 -->|MCP| Docker[Docker Engine]
Docker --> Containers[Managed Containers]
end
HyperCode V2.0 employs a strict 3-tier network segmentation strategy to enforce least-privilege access:
- frontend-net: Public-facing zone. Contains
nginx(Prod) or exposed ports fordashboardandgrafana. - backend-net: Internal application zone. Hosts
hypercode-core, Agent Swarm, and monitoring collectors. No direct external access. - data-net: Restricted data zone. Hosts
postgresandredis. Only accessible byhypercode-coreand authorized data agents.
All inter-service communication is strictly controlled via Docker internal networks.
We maintain comprehensive documentation to ensure clarity and traceability.
- Quickstart: QUICKSTART
- Mission API: MISSION.API
- Manifesto: HYPERCODE-MANIFESTO
- Getting Started: Onboarding Guide
- Technical Deep Dive: Architecture
- Integration: API Reference | MCP Integration
- Operations: Runbook | Security Threat Model
- Quality Assurance: Traceability Matrix | Benchmarks
- Governance: ADR-001 | Changelog
# Core tests (stable)
pytest -v -m "not experimental"
# All tests (including WIP)
pytest -v- Endpoint:
POST /orchestrator/mission/{mission_id}/report - Payload: JSON report including
agent_idand structured results - Artifacts: Saved under
reports/health_check_{agent_id}_{timestamp}.json - Audit: Retrieved via
GET /orchestrator/{mission_id}/audit
- Open
http://localhost:8088/ - Use Health Check Report section to submit reports
- Toggle Auto-Refresh to poll audit entries with backoff
- Last refresh timestamp and status shown inline
We welcome contributions from everyone, especially neurodivergent developers! Please read our Contributing Guide to get started.
This project is licensed under the MIT License.
Built with WelshDog + BROski ππ