Welcome to SO1
SO1 is an AI-powered automation platform that combines intelligent agents with n8n workflow automation to help engineering teams build, deploy, and operate software more efficiently.Quick Start
Get up and running with SO1 in under 5 minutes
Agents
Explore 20 specialized AI agents for every workflow
Runbooks
Operational guides for common tasks and incidents
API Reference
Integrate SO1 into your applications
What is SO1?
SO1 combines three powerful systems:AI Agents
20 specialized agents organized into 7 domains: Orchestration, Automation, Engineering, DevOps, Documentation, Prompts, and Incident Response. Each agent is trained for specific tasks within the SO1 ecosystem.
n8n Workflows
Visual workflow automation powered by n8n with 400+ integrations. Connect GitHub, Slack, databases, APIs, and more.
The SO1 Automation Factory
Figure 1 — SO1 Automation Factory. VERITAS supplies prompts bidirectionally to the SO1 agent fleet; agents orchestrate n8n workflows. All agent outputs pass through FORGE stage gates before reaching the Hono-based control plane API.
Trace ID Requirement Rationale Test Case SO1-INTRO-001The automation factory shall coordinate agents, workflows, and prompts through a unified control plane Prevents siloed execution and enables end-to-end traceability TC-SO1-INTRO-001SO1-INTRO-002FORGE gates shall be the sole mechanism for promoting agent outputs to production Ensures consistent quality across all agent domains TC-SO1-INTRO-002
Key Features
20 Specialized AI Agents
20 Specialized AI Agents
Purpose-built agents for:
- Orchestration: Task decomposition and multi-agent coordination
- Automation: n8n workflow design and webhook integrations
- Engineering: Hono backends, Next.js frontends, TypeScript types
- DevOps: Railway deployments, GitHub Actions, pipeline auditing
- Documentation: Mintlify docs, API specs, operational runbooks
- Prompts: Prompt engineering, chain design, fragment management
- Incident: Incident command, triage, and postmortem analysis
FORGE Stage-Gated Execution
FORGE Stage-Gated Execution
Every agent operates within the FORGE framework:
- Stage 0: Planning - Requirement analysis
- Stage 1: Design - Architecture decisions
- Stage 2: Implementation - Code/workflow generation
- Stage 3: Documentation - Specs and guides
- Stage 4: Validation - Testing and audit
- Stage 5: Deployment - Production release
Veritas Prompt Library
Veritas Prompt Library
A curated library of 255 prompts that agents consume and produce:
- Reusable prompt fragments
- Versioned and validated
- Bidirectional integration with agents
Schema-Validated Outputs
Schema-Validated Outputs
All agent outputs conform to defined JSON schemas:
- n8n workflow JSON
- OpenAPI specifications
- Veritas prompt drafts
Agent Domains
| Domain | Agents | Purpose |
|---|---|---|
| Orchestration | 2 | Multi-agent coordination and FORGE gate enforcement |
| Automation | 3 | n8n workflow design, webhooks, scheduling |
| Engineering | 3 | TypeScript code generation (Hono, Next.js, shared types) |
| DevOps | 3 | CI/CD pipelines, Railway deployments, security audits |
| Documentation | 3 | Mintlify docs, API specs, runbooks |
| Prompts | 3 | Prompt optimization, chains, and fragments |
| Incident | 3 | Incident response, triage, and analysis |
Getting Help
Discord Community
Join our community for help and discussions
GitHub Issues
Report bugs or request features