A.1 Core Settings

A.2 Neo4j Settings

A.3 PostgreSQL Settings

A.4 Redis / Cluster Settings

A.5 Collection Settings

A.6 Enterprise Settings

A.7 Health Endpoints

MeshOptixIQ exposes four health check endpoints. None of them require an API key, making them safe for use with load balancer health checks, Kubernetes probes, and Prometheus scraping.

GET /health

curl http://localhost:8000/health

{"status": "ok", "version": "0.9.0"}

GET /health/ready

curl http://localhost:8000/health/ready

{
  "status": "ready",
  "backend": "neo4j",
  "connected": true
}

When GRAPH_BACKEND=postgres, the response also includes:

{
  "status": "ready",
  "backend": "postgres",
  "connected": true,
  "pool_available": 8
}

Use this endpoint as the Kubernetes readiness probe. It returns HTTP 503 if the database is unreachable.

GET /health/license

curl http://localhost:8000/health/license

{
  "plan": "pro",
  "expires": "2027-01-01",
  "days_remaining": 303,
  "demo_mode": false
}

For community plan (no license key): expires and days_remaining are null.

GET /health/redis

curl http://localhost:8000/health/redis

{
  "cluster_mode": true,
  "redis_reachable": true,
  "redis_url": "redis://redis:***@6379/0"
}

The password in redis_url is masked with ***. When Redis is not configured, cluster_mode is false and redis_reachable is null.

Kubernetes probe configuration

livenessProbe:
  httpGet:
    path: /health
    port: 8000
  initialDelaySeconds: 5
  periodSeconds: 10

readinessProbe:
  httpGet:
    path: /health/ready
    port: 8000
  initialDelaySeconds: 10
  periodSeconds: 15
  failureThreshold: 3

B.0 Global Flags

B.1 meshq ingest

Parse and ingest a raw device capture file into the graph database.

meshq ingest <file> [OPTIONS]

B.2 meshq collect

Connect to live devices via SSH and collect configuration and state data.

meshq collect --inventory <file> [OPTIONS]

B.3 meshq parse

Parse a raw device capture without ingesting — useful for validating parser output.

meshq parse <file> [--vendor VENDOR] [--output json|table]

Outputs the parsed NetworkFacts structure. Exit code 0 on success, 1 if parsing fails.

B.4 meshq status

Show the current status of the API, graph backend, and license.

meshq status [--json]

Queries /health, /health/ready, and /health/license and prints a human-readable summary or JSON object.

B.5 meshq version

Print the CLI and API version information.

meshq version

meshq CLI   v0.9.0
API server  v0.9.0 (connected)
Python      3.12.4

B.6 meshq license

Display current license information.

meshq license [--json]

Plan:            pro
Expires:         2027-01-01
Days remaining:  303
Demo mode:       false

Returns exit code 2 if the license is expired.

B.7 meshq login

Authenticate and store a Personal Access Token (PAT) in the local credentials file.

meshq login [OPTIONS]

In interactive mode, prompts for the API URL and your API key or OIDC token. Stores the result in ~/.meshoptixiq/credentials (mode 0600). Subsequent CLI commands read credentials from this file automatically.

B.8 meshq logout

Remove stored credentials from the local credentials file.

meshq logout

Deletes the ~/.meshoptixiq/credentials file. After logout, every CLI command requires --api-url and --api-key flags or the corresponding environment variables.

B.9 meshq export

Export graph data in various formats.

meshq export [OPTIONS]

The ansible format produces a dynamic inventory JSON compatible with ansible-inventory --list. Devices are grouped by vendor, os_version prefix, and a firewalls group for devices with firewall rules.

# Pipe directly to Ansible
ansible-playbook -i <(meshq export --format ansible) site.yml

B.10 meshq sync

Synchronise graph data with external systems.

meshq sync --target <target> [OPTIONS]

Requires the [integrations] extra: pip install 'meshoptixiq-network-discovery[integrations]' and the NETBOX_URL / NETBOX_TOKEN environment variables. See §13.8 for full workflow.

C.1 Network Device Parsers

C.2 Firewall Policy Parsers

D.1 Topology Queries

D.2 Endpoint Queries

D.3 Addressing Queries

D.4 Hygiene Queries

D.5 Summary Queries

D.6 Inventory Queries

D.7 Firewall Queries

D.8 Routing Queries Pro+

D.9 InfiniBand & GPU Queries Enterprise

D.10 Interface Metrics Queries Pro+

D.11 NCCL / Training Job Queries Enterprise

F.1 Topology Tools

F.2 Endpoint Tools

F.3 Addressing Tools

F.4 Firewall Tools

F.5 Hygiene Tools

F.6 Summary & Inventory Tools

F.7 Administrative Tools

F.8 MCP Resources

Resources expose read-only structured data to the MCP host. Unlike tools, resources are fetched on demand by the AI agent without requiring a function call.

F.9 MCP Prompts

Prompts are pre-built instruction templates that guide the AI model through common network analysis workflows.