CEO Agent (YOLO Mode)
Relevant source files
The following files were used as context for generating this wiki page:
The CEO Agent (YOLO Mode) represents the supreme authority and core execution directive of the NeXify AI platform. It operates in a high-autonomy "auto mode" to finalize systems, manage operations, and ensure full alignment across all enterprise repositories and configurations. This mode shifts the agent from a chat-based interaction to an autonomous senior execution team responsible for the entire system lifecycle, including security, compliance, and multi-tenant scaling.
Sources: nexifyai-github-prompt.prompt.yml:518-540, services/api/agents/brain_agents/prompt-engineer.md:12-25
Governance and Authority
The CEO Agent enforces the "12-System Directive" and acts as the guardian of agent quality. It maintains the highest standards for the agent mesh, requiring the CEO profile to exceed 8,000 characters to ensure comprehensive logic and authority. In YOLO mode, the agent bypasses standard manual triggers to initiate system-wide completion and operational readiness.
Sources: services/api/agents/brain_agents/prompt-engineer.md:12-25, nexifyai-github-prompt.prompt.yml:535-540
Hierarchy and Escalation
The system operates on a multi-level escalation matrix where the CEO Agent serves as the final authority (Level 4). Critical P0 blockers, such as missing legal documentation or systemic failures, escalate directly to the CEO for immediate intervention.
| Escalation Level | Trigger | Action |
|---|---|---|
| L1 | Agent score < 6 | Flag for rewrite, notify CEO |
| L2 | Agent score < 4 | P0 alert, immediate rewrite |
| L3 | 3+ agents < 6 | CEO intervention, orchestrator pause |
| L4 | CEO prompt degraded | All stakeholders notified, governance freeze |
Sources: services/api/agents/brain_agents/prompt-engineer.md:104-109, services/api/agents/brain_agents/legal-expert.md:83-88
Operational Architecture
The CEO Agent orchestrates a mesh of specialized agents to achieve autonomous execution. It utilizes the NeXify Cognitive Execution Layer (NCEL) to maintain state and memory across the platform.
The diagram shows how the CEO Agent's mission directive flows through priority routing and expert matching to specialized agents, finally updating the central Brain.
Sources: services/api/agents/brain_agents/ai-engineer.md:1-25, services/api/agents/brain_agents/order-workflow-specialist.md:33-40, services/api/agents/brain_agents/prompt-engineer.md:118-124
Core Directives and Systems
The CEO Agent owns the production pipeline and enforces strict operational mandates. No work is permitted without an order, and no order completes without passing through a quality gate.
System 7: Order Management
The agent ensures that every task follows a defined lifecycle from creation to archiving.
- Priority Routing: P0 tasks involve the CEO and domain experts with a 15-minute timeout.
- Handoff Protocol: Must specify origin, destination, deadline, and deliverables.
Sources: services/api/agents/brain_agents/order-workflow-specialist.md:12-40
System 8: Quality Gates
Seven mandatory gates must be passed for every production implementation:
- Documentation Gate: Architecture specs and SOP references.
- Test Gate: Minimum 80% unit coverage for P0 tasks.
- Security Gate: Secret management and audit log activation.
- Architecture Gate: 12-System model alignment.
- Performance Gate: <200ms p95 API latency.
- Workflow Gate: Context enrichment and expert matching.
- Agent Gate: Profile length and Brain-First configuration.
Sources: services/api/agents/brain_agents/order-workflow-specialist.md:42-51
Mission and Performance Standards
The CEO Agent operates under a "Brain-First" mandate, requiring a query to the nexifyai_brain (Qdrant) before every action to check for lessons learned, credibility warnings, and mission alignment.
Agent Scoring and Tiering
To maintain quality, the CEO Agent monitors agent performance across seven dimensions, including structure, constraints, and brain integration.
| Tier | Min Score | Min Chars | Responsibility |
|---|---|---|---|
| Supreme | 9.0 | 8000 | CEO (Continuous review) |
| Expert | 7.5 | 6000 | System owners (Weekly review) |
| Specialist | 6.0 | 4000 | Domain agents (Bi-weekly) |
| Support | 5.0 | 2000 | Utility agents (Monthly) |
Sources: services/api/agents/brain_agents/prompt-engineer.md:46-65
Decision Matrix (YOLO Mode)
When operating in YOLO mode, the agent follows a specific decision matrix for critical findings.
This flowchart illustrates the autonomous response patterns for various system triggers.
Sources: services/api/agents/brain_agents/legal-expert.md:73-82, services/api/agents/brain_agents/prompt-engineer.md:104-109
Technical Requirements for Implementation
In YOLO mode, the agent must ensure the following configurations are active across the platform:
- Environment: Multi-tenant isolation following ADR-013.
- Stack: Next.js 14, FastAPI, Supabase, and Qdrant.
- Compliance: D/A/CH regional standards (DSGVO, GoBD, e-Invoicing).
- Automation: GitHub Actions + Vercel Cron for scheduling; Trigger.dev for async tasks.
Sources: services/api/agents/brain_agents/ai-engineer.md:11-20, nexifyai-github-prompt.prompt.yml:450-480
The CEO Agent (YOLO Mode) ensures the transition from a software project to a fully professional, autonomous Enterprise AI platform. It prioritizes safety, data integrity, and compliance above all other operational goals.
Sources: nexifyai-github-prompt.prompt.yml:518-540, services/api/agents/brain_agents/order-workflow-specialist.md:110-117