Connector catalog
Every connector is a small YAML spec: how to connect, what to collect, and how far automated resolution is allowed to go. Pick the ones that match your estate; the platform runs them under one timeline, one runbook library and one autonomy model.
Monitoring & Observability
The open-source core of Kuant NMS. These connectors are both managed by the platform and used by it: the platform deploys and operates the stack, then reads its signals to drive resolution.
- Zabbix: The detection engine of Kuant NMS: agent, SNMP, IPMI and API-based monitoring across servers, network devices and applications.
- Grafana: Dashboards and single-pane visualisation over every data source the platform manages, provisioned as code.
- Prometheus: Scrape-based metrics for Kubernetes and exporter-rich estates, federated into the platform's VictoriaMetrics store.
- Loki: Log aggregation with label-based indexing: the evidence store the AI layer quotes when it explains an incident.
- Tempo: Distributed tracing for request-level evidence: which service in the chain actually failed.
- Alertmanager: Alert routing, grouping and silencing for Prometheus-native estates, unified with Zabbix alerts in one pipeline.
Databases
PostgreSQL and the components of a production-grade HA cluster. Monitored to a 99.99 percent availability target, with failover, pooling and backup verification handled as first-class checks.
- PostgreSQL: Deep monitoring and staged remediation for PostgreSQL, run to a 99.99 percent availability target.
- Patroni: HA orchestration for PostgreSQL: automatic failover supervised, verified and reported.
- etcd: The consensus store under Patroni and Kubernetes: small, critical and monitored accordingly.
- PgBouncer: Connection pooling health: the layer that decides whether application connections wait, fail or flow.
- HAProxy: Traffic routing for database and application tiers: backend health, failover behaviour and latency, observed in one place.
- pgBackRest: Backup and point-in-time recovery, treated as a monitored service with rehearsed restores.
Infrastructure as Code
Terraform-based provisioning under the same control plane: plan visibility, drift detection and policy scanning before anything reaches production.
- Terraform / OpenTofu: Provisioning under supervision: plan visibility, drift detection and safe applies for the platform's own infrastructure.
- Terragrunt: DRY orchestration across many Terraform modules and environments, with run ordering the platform understands.
- Checkov: Policy scanning for IaC: misconfigurations caught in the repo, not in the audit.
- tfsec: Fast static analysis for Terraform, run as a pre-merge gate alongside Checkov.
Configuration Management
Ansible is the hands of Kuant NMS. Every automated fix the platform proposes or executes is an auditable playbook run, not an SSH session.
- Ansible: The execution layer of Kuant NMS: every automated fix is a versioned, idempotent, auditable playbook run.
- AWX: Web-based execution, scheduling and RBAC for the Ansible layer: who may run what, where, visible to everyone.
- Ansible-Lint: Quality gate for the runbook library: a fix that will run unattended is held to production code standards.
- Molecule: Test harness for runbooks: a remediation is proven in a disposable instance before it touches production.
CI/CD & Containers
Pipelines, registries and the container platforms they deploy to. The platform watches the delivery chain end to end and can roll back what it can prove is broken.
- GitLab: Source control and CI under the same operational lens: pipeline health, runner capacity and the platform's own GitOps source.
- Jenkins: Legacy CI estates kept observable: queue health, agent capacity and failing jobs, without a migration project first.
- Docker: Container hosts outside Kubernetes: standalone Docker and Compose estates, monitored and repaired per host.
- Kubernetes: Cluster operations with staged autonomy: from watching pods crash to fixing the reasons they crash.
- Helm: Release state as a monitored surface: what is deployed, what drifted and what a rollback would return to.
- Argo CD: GitOps as the enforcement layer: git is the desired state, and divergence is an incident with a diff.
- Harbor: Private registry health and supply chain hygiene: what is stored, what is vulnerable, what is actually pulled.
Agentic AI
The self-hosted AI layer that reasons over alerts, evidence and runbooks. All inference stays on your infrastructure: no telemetry leaves the estate.
- Ollama: Local LLM serving for the AI resolution layer: models run on your hardware, prompts never leave the estate.
- vLLM: High-throughput inference serving where the AI layer needs concurrency: batch triage, log summarisation, report generation.
- LangChain: The framework layer for AI pipelines: retrieval, tool calling and structured output over estate evidence.
- LangGraph: The agent runtime of the resolution layer: stateful graphs that investigate, decide and act within guardrails.
- LlamaIndex: Retrieval over operational knowledge: runbooks, past incidents, vendor docs and configs, indexed and cited.
- n8n: Workflow automation for the connective tissue: notifications, approvals, escalations and cross-system handoffs.
- Qdrant: The vector store under the AI layer: incident memory and runbook retrieval, operated like any other database.
Cloud & Virtualisation
Private cloud platforms and the hypervisors underneath them: capacity, cluster health, storage and live workload placement.
- Proxmox VE: Private cloud operations on Proxmox: cluster health, guest placement, storage and backup, with staged remediation.
- OpenStack: Large private clouds: API health, capacity and tenant workloads across Nova, Neutron, Cinder and friends.
- KVM / QEMU: Bare libvirt hosts without a management layer: monitored guests, storage pools and live migration support.
- Ceph: Distributed storage under the private cloud: health, capacity, recovery and the discipline of never losing data.
Firewall & Security
Firewalls, IDS/IPS and host security. Detection feeds the same resolution pipeline as operations: a security event becomes a ticket, an investigation and a contained host.
- pfSense: Perimeter and branch firewalls: health, VPN state, rule hygiene and config backup, managed at fleet scale.
- OPNsense: The same fleet discipline for OPNsense estates: API-first monitoring, config tracking and staged remediation.
- Suricata: Inline and passive IDS/IPS: signature-based detection feeding the same incident pipeline as operations.
- Zeek: Network metadata at depth: connection logs, DNS, TLS and file activity for investigation and hunting.
- Wazuh: Host security telemetry: file integrity, vulnerability state, log-based detection and compliance posture per host.
- WireGuard: Site-to-site and admin VPN mesh: handshake health, key rotation and peer lifecycle as managed configuration.
Wi-Fi & Network Access
The access edge: wireless, authentication and address management for offices, campuses and branch estates.
- OpenWRT: Branch and campus Wi-Fi on open firmware: fleet configuration, radio health and client experience metrics.
- FreeRADIUS: 802.1X and network authentication: the service that locks everyone out when it fails, watched accordingly.
- Kea DHCP: Modern DHCP with an API: pool exhaustion forecast, HA state and lease anomalies before users notice.
Identity & Directory
Directories, SSO and domain services. The platform watches replication, expiry and authentication health, the failures users notice first.
- OpenLDAP: Directory backbone: replication, search performance and the integrity of the data everything else authenticates against.
- FreeIPA: Integrated identity for Linux estates: directory, Kerberos, DNS and CA health as one monitored system.
- 389 Directory Server: Enterprise directory at scale: replication topology, performance counters and disk-backed integrity checks.
- Keycloak: SSO and identity federation: login health, token issuance and the realm configuration everything depends on.
- SSSD: The client side of identity: cached credentials, domain reachability and login health on every Linux host.
- Samba AD DC: Open-source Active Directory: DC replication, SYSVOL consistency, FSMO roles and DNS, monitored like the real thing.
Mail & Collaboration
Self-hosted mail estates: queues, delivery, reputation and hygiene, monitored so mail problems are caught before customers report them.
- Postfix: SMTP transport: queues, deferrals, TLS and the reputation signals that decide whether your mail is delivered.
- Dovecot: IMAP and mailbox storage: login health, index integrity and quota pressure across the mailbox estate.
- Rspamd: Spam filtering and DKIM signing: catch rates, learning health and the signing that protects your domain reputation.
- ClamAV: Malware scanning in the mail path: signature freshness and scan throughput without mail delays.
- Roundcube: Webmail front end: login health, session errors and the upgrade hygiene of an internet-facing PHP application.
- Mailcow: The full dockerised mail stack as one connector: every container, queue and certificate under a single health model.
IT Service Management
Ticketing, assets and source-of-truth inventory. Alerts become tickets automatically, and resolved incidents close their own loop.
- GLPI: Tickets and assets closing the loop: every incident the platform works becomes an auditable service record.
- Zammad: Modern helpdesk integration: incidents become conversations with full context, across email, chat and web.
- osTicket: Lightweight ticketing kept in the loop: API-driven ticket creation for estates that already run osTicket.
- Snipe-IT: Hardware asset truth: what exists, who has it, and whether monitoring coverage matches the asset register.
- NetBox: The source of truth: sites, racks, devices, IPs and circuits that every other connector reads for context.
Commercial estates
Most estates are not purely open source. These connectors bring the commercial infrastructure you already own, firewalls, switches, hypervisors, directories and ITSM, into the same catalog and the same resolution pipeline.
- FortiGate: Fortinet estates under the same pipeline: fleet health, tunnel state, config diffs and staged remediation over the REST API.
- Sophos Firewall: Sophos XGS estates: health, VPN and rule change visibility through the XML and REST APIs.
- Cisco IOS / SG Switches: Campus and data center switching: port health, config compliance and change control over SSH and SNMP.
- VMware vSphere: Existing vSphere estates: cluster capacity, datastore pressure, snapshot debt and guest health through vCenter.
- Microsoft AD / Entra ID: The directory most estates actually run: DC health, replication, account hygiene and hybrid sync into Entra ID.
- ServiceNow: Enterprise ITSM integration: platform incidents flow into your existing ServiceNow process, evidence included.
Home
Contact: support@kuant.co · +91 9716901521 · Gurugram, India
