Multi-Agent Planning (MAP)

Multi-Agent Planning (MAP) is AxonFlow's orchestration layer for coordinating multiple AI agents to accomplish complex tasks. MAP breaks down user requests into subtasks, assigns them to specialized agents, and aggregates results—all while enforcing governance policies at every step.

How It Works

User Request: "Plan a business trip to Mumbai"
                    │
                    ▼
          ┌─────────────────┐
          │ Planning Engine │  ← Analyzes intent, selects agents
          └─────────────────┘
                    │
        ┌───────────┼───────────┐
        ▼           ▼           ▼
   ┌─────────┐ ┌─────────┐ ┌─────────┐
   │ Flight  │ │ Hotel   │ │ Policy  │  ← Specialist agents
   │ Agent   │ │ Agent   │ │ Checker │
   └─────────┘ └─────────┘ └─────────┘
        │           │           │
        └───────────┴───────────┘
                    │
                    ▼
          ┌─────────────────┐
          │ Workflow Engine │  ← Executes steps, aggregates results
          └─────────────────┘
                    │
                    ▼
            Aggregated Response

Key Components

Planning Engine

The Planning Engine analyzes incoming requests and generates execution plans:

• Query Analysis: Detects intent and required capabilities
• Agent Selection: Chooses appropriate agents based on capabilities
• Step Generation: Creates execution steps with dependencies
• Domain Templates: Uses domain-specific templates (travel, healthcare, finance)

Agent Registry

The Agent Registry manages available agents:

Mode	Description	Edition
File-based	Agents defined in YAML configuration files	Community
Database-backed	Agents stored in PostgreSQL with CRUD API	Enterprise
Hybrid	Both file-based and database agents (DB takes priority)	Enterprise

Workflow Engine

The Workflow Engine executes plans:

• Step Execution: Runs each step type (llm-call, connector-call, etc.)
• Dependency Management: Ensures steps run in correct order
• Parallel Execution: Runs independent steps simultaneously
• Template Variables: Passes data between steps via {{input.key}} and {{steps.name.output}}

Step Types

MAP supports five step types for different operations:

Step Type	Description	Use Case
llm-call	Invoke an LLM provider for inference	Text generation, analysis
connector-call	Query an MCP connector	Database queries, API calls
conditional	Branch based on conditions	Decision logic
function-call	Execute a custom function	Data transformation
api-call	Call an external HTTP API	Third-party integrations

See Step Types for detailed configuration.

Agent Types

Type	Description	Example
Specialist	Single-domain expertise	Flight search, data analysis
Coordinator	Orchestrates other agents	Trip planner, workflow manager

Execution Modes

Mode	Description	When to Use
Sequential	Steps run in order, each receiving previous output	Dependent tasks
Parallel	Independent steps run simultaneously	Multiple data sources
Conditional	Steps run based on conditions	Decision trees

Community vs Enterprise

Feature	Community	Enterprise
File-based agent registry	✅	✅
Plan generation	✅	✅
Sequential execution	✅	✅
Parallel execution	✅	✅
All 5 step types	✅	✅
Database-backed agents		✅
Agent CRUD API		✅
Agent versioning		✅
Multi-tenant isolation (RLS)		✅
Execution analytics		✅
Visual workflow builder		✅

Quick Example

1. Define an Agent

# agents/research-agent.yaml
apiVersion: axonflow.io/v1
kind: AgentConfig
metadata:
  name: research-agent
  domain: generic
spec:
  type: specialist
  description: Research and summarize information
  capabilities:
    - research
    - summarization
  llm:
    provider: openai
    model: gpt-4

2. Generate a Plan

curl -X POST http://localhost:8081/api/v1/plan/generate \
  -H "Content-Type: application/json" \
  -d '{
    "query": "Research the benefits of remote work",
    "domain": "generic"
  }'

3. Execute the Plan

curl -X POST http://localhost:8081/api/v1/plan/execute \
  -H "Content-Type: application/json" \
  -d '{
    "plan_id": "plan_abc123",
    "step_index": 0,
    "input": {
      "query": "Research the benefits of remote work"
    }
  }'

Next Steps

• Getting Started - Set up your first multi-agent workflow
• Agent Configuration - Learn the full agent YAML schema
• Step Types - Deep dive into each step type
• Planning Patterns - Common orchestration patterns

Enterprise Features

For database-backed agents, CRUD APIs, versioning, and analytics:

• Enterprise MAP Guide - Full enterprise documentation
• Request a Demo - See MAP in action