Core Architecture
Human-in-the-Loop Native
Human oversight isn't bolted on after the fact. Approval nodes, review queues, and escalation paths are built into the workflow engine from the ground up.
Approval Nodes
Require human sign-off at any point in a workflow before proceeding.
Review Queues
Route AI outputs to the right reviewer with priority and assignment rules.
Escalation Paths
Define fallback chains when approvals aren't granted or deadlines pass.
Human Task Assignment
Assign specific tasks to specific people based on role, expertise, or availability.
Timeout & Fallback
Configurable timeouts with automatic escalation or alternative handling.
AI Capabilities
Multi-Framework Agent Support
Bring your own framework. AICtrlNet integrates with the major AI agent frameworks so you can use the tools your team already knows.
5 Framework Integrations
LangChain, AutoGPT, CrewAI, AutoGen, and Semantic Kernel supported out of the box.
41 Pre-built Agents
33 AI agents and 8 specialized agents ready to deploy in your workflows.
Custom Agent Creation
Build your own agents with the Python SDK or the visual builder.
Agent Collaboration
Agent pods enable multi-agent collaboration within a single workflow.
Build
Visual + Code Workflow Engine
Build workflows visually with drag-and-drop or programmatically with Python. Conditional logic, loops, branching, error handling, and retry built in.
Visual Workflow Builder
Drag-and-drop interface for building workflows without writing code.
183 Workflow Templates
Pre-built templates for common patterns across industries.
Conditional Logic
Branching, loops, and conditional execution based on data or human decisions.
Error Handling & Retry
Configurable error handling, retry policies, and dead-letter queues.
Connect
Protocols & Integrations
Native support for MCP (Model Context Protocol) and A2A (Agent-to-Agent), plus 27 adapters for connecting to the services your team already uses.
Adapters
Salesforce
HubSpot
Slack
Teams
Email
PostgreSQL
MongoDB
S3
GCS
OpenAI
Anthropic
n8n
Zapier
Make
+13 more
MCP Native
Model Context Protocol support for standardized AI model interactions.
Google A2A
Agent-to-Agent protocol for cross-platform agent communication.
Platform Bridges
Integrate with n8n, Zapier, and Make for existing automation workflows.
Govern
AI Governance & Compliance
Production AI needs oversight. Audit trails, bias detection, and explainability reports give your team confidence that AI is operating within bounds.
Audit Trails
Complete record of every decision, approval, and AI action across workflows.
Bias Detection
ML-powered analysis to flag potential bias in AI outputs before they reach production.
Explainability Reports
Human-readable explanations of AI decisions for compliance and review.
Compliance Frameworks
Built-in support for HIPAA, GDPR, and SOC2 compliance requirements.
Develop
Developer Experience
Python SDK, REST API, CLI tools, and Docker-first deployment. Build locally, deploy anywhere.
$ docker pull bodaty/aictrlnet-community
$ docker run -p 8000:8000 bodaty/aictrlnet-community
$ pip install aictrlnet
$ aictrlnet init my-project
Python SDK
Build and manage workflows programmatically with a clean Python API.
REST API
Full REST API for integration with any language or platform.
CLI Tools
Command-line tools for project scaffolding, deployment, and management.
Docker & Kubernetes
Container-native deployment with Docker Compose and Kubernetes support.
Ready to build?
Get started with the Community Edition or compare editions to find the right fit for your team.