Tensorify LogoTensorify
Tensorify LogoTensorify
Log inGet Started

System Info

Query system resources on the machine where the Tensorify CLI runner is running — CPU, memory, disk, processes, network interfaces, open ports, and uptime.

When to Use

Use System Info when you need to:

  • Monitor CPU, memory, or disk usage on the runner host
  • List running processes or open ports for health checks
  • Build alerting workflows when resource thresholds are exceeded
  • Give an AI Agent visibility into the runner machine via the Tools handle

Best with the CLI runner (tensorify runner start) — queries report your machine's real resources. In managed mode, System Info runs but reports the sandbox container's resources, not your infrastructure. Use CLI execution mode when monitoring your own servers.

Inputs

HandleTypeDescription
inputanyUpstream data for {{ input.field }} template references in settings fields.

Output

HandleTypeDescription
resultobjectSystem query result — fields present depend on the query type

The result (accessed as {{ system_info.* }}) shape includes these fields (not all are present for every query):

FieldTypeDescription
cpu_percentnumberOverall CPU utilization (%) — cpu query
memory_used_gbnumberUsed memory in GB — memory query
memory_total_gbnumberTotal memory in GB — memory query
disk_free_gbnumberFree disk space in GB — disk query
disk_total_gbnumberTotal disk space in GB — disk query
processesarrayList of process objects — processes query
portsarrayList of open connection objects — ports query
uptime_hoursnumberHours since last boot — uptime query
networkarrayList of network interface objects — network query
successbooleanWhether the query succeeded — all queries

Example outputs by query type:

// cpu
{ "cpu_percent": 23.5, "success": true }

// memory
{ "memory_used_gb": 10.245, "memory_total_gb": 32.0, "success": true }

// disk
{ "disk_free_gb": 120.5, "disk_total_gb": 500.0, "success": true }

// processes
{ "processes": [{ "pid": 1234, "name": "python3", "cpu_percent": 2.5 }], "success": true }

// ports
{ "ports": [{ "port": 8080, "status": "LISTEN", "pid": 9012 }], "success": true }

// uptime
{ "uptime_hours": 72.5, "success": true }

// network
{ "network": [{ "interface": "eth0", "addresses": [{ "address": "192.168.1.100" }] }], "success": true }

Settings

Query

SettingTypeDefaultDescription
queryenumcpuQuery type: cpu, memory, disk, processes, network, ports, or uptime.
allowed_operationsmulti-selectall query typesWhich query types are available when this node is connected to an AI Agent Tools handle. Excluded from the agent tool schema.
filterstring—Filter results. For disk: mount path (default /). For processes: process name substring. For ports: port number. Not used for cpu, memory, network, or uptime. Supports {{ }} bindings.

Example

Canvas

Disk usage health check:

  1. Add a System Info node.
  2. Set Query to disk.
  3. Set Filter to /.
  4. Add an If node checking {{ system_info.disk_free_gb }} / {{ system_info.disk_total_gb }} < 0.1.
  5. On the true branch, send an alert via Send Email or HTTP Request.

Process monitoring:

  • Query: processes
  • Filter: tensorify

Access results:

{{ system_info.processes }}
{{ system_info.success }}

TSL

import manual from @tensorify/manual-trigger:1.0.0
import sysinfo from @tensorify/system-info:1.0.0

node trigger @tensorify/manual-trigger:1.0.0 {}

node disk @tensorify/system-info:1.0.0 {
    query = "disk"
    filter = "/"
}

trigger.payload -> disk.input

Error Handling

System Info has an On Error control output branch. Enable it via the "Error output handle" toggle at the bottom of the settings panel.

The error branch fires when:

  • A query fails due to a runtime error (e.g. invalid mount path for disk)
  • A query type is rejected because it is not in Allowed Operations (Agent) when used as an agent tool

Connect the error branch to a fallback notification or logging action. You can also check {{ system_info.success }} on the main output path.

Common Gotchas

  • Managed mode reports sandbox resources: In managed mode, System Info reports the sandbox container's CPU, memory, and disk — not your machine's. Deploy with CLI mode to monitor your own infrastructure.
  • CPU sampling delay: The cpu query includes a 1-second sampling interval for accurate percent readings — expect a brief pause.
  • Filter usage varies by query: Use Filter for mount paths (disk), process name substrings (processes), and port numbers (ports). It is ignored for cpu, memory, network, and uptime.
  • Ports query may require elevated permissions: Listing open network connections can fail on some systems without sufficient process permissions.
  • Agent permissions: Restrict Allowed Operations (Agent) to cpu and memory if you do not want agents inspecting processes or ports.

See Also

  • AI Agent — connect System Info as a monitoring tool for agents
  • Shell — run commands on the runner host
  • If — branch on resource thresholds
  • Send Email — send alerts when thresholds are exceeded
  • Deploying Workflows — configure CLI execution mode
On this page