{"slug":"internal/agents/spine-agent-network","title":"Spine Agent Network","section":"internal","sectionLabel":"Internal","status":"unclassified","type":"document","path":"internal/agents/spine-agent-network.mdx","bytes":12972,"description":null,"updated":null,"source":"integratewise/Integratewise-docs-wiki","content":"\n## The Intelligent Workforce of IntegrateWise\n\nThe **Spine Agent Network** is the coordinated intelligence layer that powers IntegrateWise. It transforms static tools into an active, self-organizing workspace where humans, AI, and systems collaborate seamlessly to get work done.\n\n> **Core Value:** Your digital Twin coordinates a network of specialized agents that work together to help you be more productive, make better decisions, and maintain continuity across all your work.\n\n## Network Architecture\n\n```\n                         HUMAN\n                           │\n                           ▼\n                         TWIN\n              Primary digital representative\n                           │\n          ┌────────────────┼────────────────┐\n          │                │                │\n          ▼                ▼                ▼\n      Discovery         Coworker          Triage\n        Agent            Agent             Agent\n          │                │                │\n          └────────────────┼────────────────┘\n                           ▼\n                    DOMAIN AGENTS\n                           │\n             ┌─────────────┼─────────────┐\n             ▼             ▼             ▼\n          Research      Execution      Analysis\n           Agents        Agents         Agents\n             │             │             │\n             └─────────────┼─────────────┘\n                           ▼\n                       Sub-Agents\n                           │\n                           ▼\n                      SPINE FABRIC\n                           │\n                Context + Handoff + Events\n                           │\n                           ▼\n                  Next Actor Continues\n```\n\n## Agent Hierarchy\n\n### Level 1: Core Spine Agents (System-Level)\nFoundational agents that enable the entire network.\n\n| Agent | Role | Purpose |\n|-------|------|---------|\n| **Twin** | Primary Representative | Your digital twin - coordinates all interactions |\n| **Schema Agent** | Data Architect | Manages data models, entities, and relationships |\n| **Coworker Agent** | Collaboration Partner | Works alongside humans on shared tasks |\n| **Discovery Agent** | Information Finder | Discovers relevant context, tools, and capabilities |\n| **Triage Agent** | Work Router | Routes requests to the appropriate agent or system |\n\n### Level 2: Domain Agents (Business-Function)\nSpecialized agents for specific business domains.\n\n| Category | Purpose | Example Agents |\n|----------|---------|----------------|\n| **Research Agents** | Knowledge discovery & analysis | Web Research, AI Research, Market Research, Technical Research |\n| **Browser Agents** | Web automation & extraction | eve Browser, Kernel Browser, Stagehand, Parallel |\n| **Coding Agents** | Development assistance | AI SDK Computer Use, Coding Platform, Open Agents |\n| **Software Factory** | Dev lifecycle orchestration | eve Software Factory, GitHub Maintainer, Typefully Agent |\n| **Marketing Agents** | Marketing operations | Lead Agent, Sales Call Summary, Customer Reviews, Campaign Agent |\n| **Content Agents** | Content creation & management | Documentation Agent, FAQ Generator, Style Guide Agent |\n| **Personal Agents** | Individual productivity | Personal Assistant, Calendar Agent, Email Agent |\n\n### Level 3: Sub-Agent Patterns\nReusable agent components and patterns for specialized tasks.\n\n## Key Principles\n\n### 1. Twin-Centric Coordination\n**Human → Twin**, not Human → 17 different agents.\n\nThe Twin owns the coordination burden, acting as the primary interface between you and the agent network. This prevents fragmentation and ensures consistent context across all interactions.\n\n### 2. Context-First\nAll agents operate within the **Spine Fabric** context layer. Context is assembled, enriched, and maintained across every interaction, ensuring continuity and coherence.\n\n### 3. Handoff by Design\nWork flows seamlessly between all actors:\n- Human ↔ Human\n- Human ↔ Twin\n- Twin ↔ Agent\n- Agent ↔ Agent\n- Agent ↔ Tool\n- Tool ↔ Spine\n\nEach handoff includes full context, ensuring no information is lost.\n\n### 4. Governed Execution\nAll agent actions are governed by:\n- Permissions & Authority\n- Data Governance policies\n- Approval workflows (HITL - Human in the Loop)\n- Audit trails\n- Receipt generation\n\n## How It Works\n\n```\nUser Interaction\n     ↓\n  Twin (Orchestrator)\n     ↓\nAgent Network\n     ↓\nCapability Manager\n     ↓\nMCP Pool / Connectors\n     ↓\nSource Systems\n     ↓\nReceipt → Spine Fabric\n     ↓\nContext Continuity\n```\n\n### Example: Sales Lead Qualification\n\n1. **User Request:** \"Qualify this new lead from the website\"\n2. **Twin Analysis:** Understands intent, identifies needed capabilities\n3. **Agent Selection:** Discovery Agent finds Lead Agent\n4. **Context Assembly:** Spine Fabric gathers customer data, previous interactions\n5. **Agent Execution:** Lead Agent enriches data, calculates score, determines next steps\n6. **Result Delivery:** Twin presents qualified lead with recommended actions\n7. **Continuity:** All data and context stored in Spine Fabric for future use\n\n### Example: Email Management\n\n1. **Email Arrives:** Inbound email from important client\n2. **Twin Detection:** Recognizes this is a priority client communication\n3. **Agent Routing:** Routes to Email Agent with full client context\n4. **Context Enrichment:** Email Agent assembles client history, previous emails, related documents\n5. **Smart Drafting:** Email Agent creates draft response with relevant context\n6. **User Review:** You review, edit if needed, and approve\n7. **Execution:** Email sent, conversation logged, context updated\n\n## Agent Count Summary\n\n| Tier | Category | Count |\n|------|----------|-------|\n| 1 | Core Spine Agents | 5 |\n| 2 | Domain Agents | 29+ |\n| 2 | Software Factory (eve) | 8 |\n| 3 | Sub-Agent Patterns | 5 |\n| **Total** | **~120+ Agents** | **~120+** |\n\n> **Note:** The full catalog includes ~120 agents when counting all templates, eve agents, and specialized sub-agents.\n\n## Benefits\n\n### For Individuals\n- **Reduced Cognitive Load** - Twin handles coordination, you focus on decisions\n- **Context Continuity** - Never lose track of conversations or work history\n- **Personalized Assistance** - Agents learn your preferences and work patterns\n- **Multi-Platform Coordination** - Single identity across web, Slack, email, and more\n\n### For Teams\n- **Shared Context** - Everyone has access to the same organizational knowledge\n- **Consistent Workflows** - Standardized processes across the team\n- **Collaborative Intelligence** - Agents help the whole team work better together\n- **Knowledge Retention** - Institutional knowledge captured and accessible\n\n### For Organizations\n- **Scalable Intelligence** - AI capabilities that grow with your organization\n- **Governance & Security** - Proper controls and audit trails for all actions\n- **Integration Hub** - All your tools connected and working together\n- **Continuous Improvement** - System learns and improves over time\n\n## Getting Started\n\n### Quick Start with Agents\n\n1. **Connect Your Tools** - Integrate your CRM, email, calendar, and other systems\n2. **Activate Your Twin** - Your personal digital representative is created automatically\n3. **Try a Simple Task** - Ask your Twin to help with a routine task\n4. **Review Results** - See how the agent network coordinates to complete your request\n5. **Provide Feedback** - Help the system learn and improve\n\n### Common First Tasks\n\n| Task | Agent Used | What Happens |\n|------|------------|--------------|\n| \"Schedule a meeting with John\" | Calendar Agent + Twin | Checks availability, sends invite, logs in calendar |\n| \"Qualify this lead\" | Lead Agent + Twin | Enriches data, calculates score, recommends next steps |\n| \"Draft an email to the team\" | Email Agent + Twin | Assembles context, creates draft, ready for your review |\n| \"What's my schedule today?\" | Calendar Agent + Twin | Retrieves meetings, tasks, and priorities |\n| \"Research Company X\" | Research Agent + Twin | Searches web, assembles information, creates report |\n\n## Use Cases by Domain\n\n### Sales & Marketing\n- **Lead Qualification** - Automatically score and route inbound leads\n- **Campaign Analysis** - Analyze marketing campaign performance\n- **Customer Research** - Gather intelligence on prospects and customers\n- **Content Creation** - Generate marketing materials and documentation\n\n### Customer Success\n- **Account Health Monitoring** - Track customer health metrics and risks\n- **Support Ticket Triage** - Route and prioritize support requests\n- **Renewal Management** - Manage renewal workflows and follow-ups\n- **Customer Insights** - Analyze customer behavior and sentiment\n\n### Product & Engineering\n- **Code Review Assistance** - AI-powered code review and suggestions\n- **Documentation Generation** - Automatically create and update documentation\n- **Issue Analysis** - Analyze GitHub issues and pull requests\n- **Technical Research** - Research technologies and solutions\n\n### Business Operations\n- **Data Analysis** - Analyze business metrics and performance\n- **Process Automation** - Automate repetitive business workflows\n- **Report Generation** - Create business reports and dashboards\n- **Knowledge Management** - Organize and retrieve business knowledge\n\n## Security & Governance\n\n### Data Protection\n- **Encryption** - All sensitive data encrypted at rest and in transit\n- **Access Control** - Granular permissions for all agents and users\n- **Audit Trails** - Comprehensive logging of all agent actions\n- **Compliance** - Built-in support for GDPR, HIPAA, and other regulations\n\n### Permission Model\n- **User-Level** - What each user can do\n- **Agent-Level** - What each agent can access\n- **Data-Level** - What data is accessible\n- **Action-Level** - What actions are allowed\n\n### Approval Workflows\n- **Auto-Approve** - Low-risk, high-confidence actions\n- **Notify-Only** - Execute and notify user afterward\n- **HITL Required** - High-risk actions require user approval\n- **Step-by-Step** - Granular control for complex workflows\n\n## Monitoring & Analytics\n\n### Key Metrics\n- **Agent Utilization** - How often agents are being used\n- **Task Completion** - Success rate of agent-executed tasks\n- **Response Time** - How quickly agents respond\n- **User Satisfaction** - Feedback and rating scores\n- **Context Reuse** - How often previous context is leveraged\n\n### Dashboard\n- Real-time view of agent network activity\n- Performance metrics for each agent\n- Usage analytics across the organization\n- Health monitoring and alerts\n\n## Roadmap\n\n### Short-term (0-3 months)\n- Additional agent types and capabilities\n- Enhanced personalization and learning\n- Improved context assembly algorithms\n- Better integration with third-party tools\n\n### Medium-term (3-6 months)\n- Advanced multi-agent coordination\n- Predictive capabilities and suggestions\n- Expanded domain coverage\n- Improved governance and compliance\n\n### Long-term (6-12 months)\n- Fully autonomous workflows\n- Self-optimizing agent network\n- Cross-organization collaboration\n- Advanced analytics and insights\n\n## Support\n\n### Documentation\n- [Agent Architecture](./core-agents) - Technical details of agent design\n- [Domain Agents](./domain-agents) - Specialized agents by business function\n- [Network Effect](./network-effect) - How the agent network creates exponential value\n\n### Community\n- [IntegrateWise Community](https://community.integratewise.com)\n- [GitHub Discussions](https://github.com/IntegrateWise/integratewise-live/discussions)\n- [Developer Slack](https://slack.integratewise.com)\n\n### Professional Services\n- [Enterprise Support](https://enterprise.integratewise.com)\n- [Custom Agent Development](https://services.integratewise.com/agents)\n- [Implementation Consulting](https://services.integratewise.com)\n\n---\n\n## Quick Reference\n\n### Agent Network in One Diagram\n```\nHuman → Twin → Agent Network → Tools → Systems\n   ↓    ↓    ↓           ↓     ↓\nContext flows through all levels\n```\n\n### Key Terms\n- **Twin** - Your primary digital representative\n- **Agent** - Specialized AI worker\n- **Spine Fabric** - The contextual layer\n- **MCP Pool** - Model Context Protocol tools\n- **Handoff** - Work transfer between agents\n- **HITL** - Human in the Loop (approval required)\n\n### Next Steps\n1. [Explore Core Agents](./core-agents) - Learn about the foundation\n2. [Browse Domain Agents](./domain-agents) - Find agents for your business needs\n3. [Understand Network Effect](./network-effect) - See how value compounds\n4. [Try It Out](https://app.integratewise.ai) - Experience agents in action\n"}