AI Gateway
Build any agent. Connect any tool.
The AI Gateway orchestrates autonomous agents using A2A protocol and connects them to external systems via MCP. The seven agents below are examples of what's possible — create your own to match your stack.
Example Agents
Seven agents to get you started
Each agent is built on the AI Gateway. Use them as-is, extend them, or build new ones — every agent speaks A2A and consumes tools through MCP.
Terraform Agent
Autonomous infrastructure planning, validation, and execution with auto-remediation.
- Autonomous plan generation and validation
- Pre-apply security and compliance checks
- Auto-remediation on failure
- Drift detection and correction
Security Agent
Continuous security scanning with compliance detection and SIEM integration.
- Trivy container vulnerability scanning
- Compliance gap detection
- SIEM integration and alerting
- OPA policy enforcement
Network Validation Agent
Batfish-powered network configuration validation before any change goes live.
- Pre-deployment network testing
- Routing and ACL verification
- Reachability analysis
- Firewall rule validation
Network Digital Map Agent
Automatic topology mapping and resource discovery across your entire cloud estate.
- Multi-cloud topology mapping
- Resource relationship discovery
- Visual network maps
- Change impact visualization
CogniNet Agent
AI-powered network cognition using Graph Neural Networks and a 108+ rule engine to analyze, score, and predict network behavior.
- What-if failure analysis and blast radius prediction
- Network Cognition Score (NCS) for executive reporting
- 7 specialized analyzers (routing, L2, security, capacity, convergence)
- Multi-vendor config parsing (Cisco IOS/NX-OS/ASA, Palo Alto, F5)
- Intent-based validation with YAML definitions
- Step-by-step remediation plans
Workflow Agent
Coordinates multi-step workflows across other agents using the A2A protocol.
- A2A-based multi-agent coordination
- Sequential and parallel step execution
- Rollback and compensation logic
- Error handling and retry policies
Architecture Reviewer (AAR)
AI-powered architecture assessment against industry best practices.
- Architecture pattern review
- Best practice validation
- Anti-pattern detection
- Optimization recommendations
Explore use cases
What these agents can automate
31 example infrastructure workflows — from Terraform drift detection to cross-cloud deployment orchestration — all built on the AI Gateway with A2A and MCP.
Automate Terraform Drift Detection with AI Agents
Continuous drift detection across every Terraform workspace, with blast-radius classification and PR-based remediation.
Read use caseContainer Image Vulnerability Scanning with AI Agents
Every container image scanned with Trivy, findings triaged by exploitability and reachability, and fix PRs opened automatically.
Read use casePre-Deployment Network Validation with AI Agents
Batfish-powered reachability and ACL testing before any network change reaches production, catching breakage before it ships.
Read use caseMulti-Cloud Topology Mapping with AI Agents
Automatic, continuous topology discovery across AWS, Azure, and GCP, with cross-cloud relationships and a live visual map.
Read use caseWhat-If Network Failure Analysis with CogniNet
Model link, device, and interface failures before they happen. CogniNet's GNN predicts blast radius, convergence time, and affected services for any proposed change or failure scenario.
Read use caseEnd-to-End Network Visibility with AI Agents
Continuous network visibility from config collection to digital twin to knowledge graph to live telemetry, orchestrated end-to-end by AI agents.
Read use caseAutomated AWS Well-Architected Review with AI
Continuous Well-Architected Framework assessment across every workload: reliability, security, cost, performance, operational excellence, and sustainability.
Read use caseIntegrations
Deep integrations via Model Context Protocol
Agents connect to your existing tools through a standardised MCP protocol for real-time data exchange, enabling validation, monitoring, ticketing, and alerting without custom glue code.
Network & Validation
Data & Observability
Ticketing & Collaboration
Security & Compliance
Cloud Providers
IaC & Source Control
A2A + MCP
A2A for agents. MCP for tools.
Agents talk to each other using the Agent-to-Agent (A2A) protocol and access external systems through Model Context Protocol (MCP). Both are open standards, so your agents stay portable and your tools stay reusable across every workflow.
Experience the Power of AI-Driven Infrastructure
