workspace/skills/radkit-remote-access/SKILL.md
Cisco RADKit — cloud-relayed remote device access, CLI execution, SNMP polling, device inventory discovery, attribute inspection. Use when accessing remote network devices through a cloud relay, running CLI on air-gapped devices, polling SNMP metrics remotely, or discovering device inventory via RADKit.
npx skillsauth add automateyournetwork/netclaw radkit-remote-accessInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
RADKIT_IDENTITY, RADKIT_DEFAULT_SERVICE_SERIAL, active RADKit service instanceRADKit provides a cloud-relayed path to on-premises devices:
NetClaw Agent --> RADKit Cloud --> RADKit Service (on-prem) --> Device (CLI/SNMP)
This is ideal for:
| Tool | Parameters | What It Does |
|------|-----------|--------------|
| get_device_inventory_names | none | List all onboarded device names from the RADKit service |
| get_device_attributes | target_device | Retrieve device details in JSON: host, type, configs, SNMP/NETCONF status, capabilities |
| exec_cli_commands_in_device | target_device, commands, timeout?, max_lines? | Execute CLI commands on a device with timeout and line-limit controls |
| snmp_get | target_device, oid(s), timeout? | Perform SNMP GET operations without CLI execution |
| exec_command | target_device, commands | Structured command execution — returns dict/list with status and truncation info |
Discovers all devices onboarded to the RADKit service. Call this first to know what devices are available.
Returns a set of device names, e.g.: {"edge-rtr-01", "core-sw-01", "dc-fw-01"}
Retrieves detailed JSON attributes for a specific device:
Safe for parallel execution across multiple devices.
Executes CLI commands on a device through the RADKit relay:
Use this for standard show commands, debug captures, and configuration inspection.
Performs SNMP GET without executing CLI:
Common OIDs:
| OID | Metric |
|-----|--------|
| 1.3.6.1.2.1.1.1.0 | System Description |
| 1.3.6.1.2.1.1.3.0 | System Uptime |
| 1.3.6.1.2.1.1.5.0 | System Name |
| 1.3.6.1.2.1.2.2.1.2 | Interface Description |
| 1.3.6.1.2.1.2.2.1.8 | Interface Operational Status |
Structured command execution that returns a dictionary or list:
When first connecting via RADKit:
get_device_inventory_names — what devices are available?get_device_attributes for each device — type, platform, capabilitiessnmp_get with sysUpTime (1.3.6.1.2.1.1.3.0) for each deviceWhen investigating an issue on a remote device:
get_device_inventory_names — find the target device nameget_device_attributes — confirm CLI access is availableexec_cli_commands_in_device with timeout and max_lines
show ip interface brief — interface statusshow ip route summary — routing table healthshow processes cpu sorted — CPU utilizationshow logging last 50 — recent syslog messagesexec_command for commands needing programmatic parsingWhen collecting metrics from remote devices:
get_device_inventory_names — target devicessnmp_get with sysDescr, sysName, sysUpTimesnmp_get with ifOperStatus for key interfacessnmp_get with ifInOctets, ifOutOctets for bandwidth trackingWhen checking health across sites served by the RADKit service:
get_device_inventory_names — all devices across sitesget_device_attributes for each — group by type and siteexec_cli_commands_in_device per device:
show version — uptime, software versionshow processes cpu | include CPU utilizationshow memory statisticssnmp_get for sysUpTime, interface counters| Skill | How They Work Together |
|-------|----------------------|
| pyats-network | RADKit for cloud-relayed access, pyATS for direct SSH — complementary paths to devices |
| pyats-health-check | RADKit provides remote device data; pyATS health-check procedures analyze it |
| pyats-troubleshoot | RADKit CLI exec for remote devices that pyATS can't reach directly |
| pyats-routing | Use RADKit to collect routing state from remote sites, analyze with routing skill |
| pyats-security | RADKit CLI for remote security audit commands (ACLs, AAA, CoPP) |
| meraki-monitoring | Meraki for cloud-managed devices, RADKit for on-prem devices behind Meraki MX |
| te-path-analysis | ThousandEyes external path + RADKit internal CLI for end-to-end troubleshooting |
| nso-device-ops | NSO for orchestrated config, RADKit for raw CLI access to same devices |
| gait-session-tracking | Record all RADKit remote access sessions in GAIT |
| servicenow-change-workflow | Gate any config changes through RADKit with ServiceNow CRs |
snmp_get over CLI for structured metrics (uptime, counters, status).RADKIT_IDENTITY — User email address for RADKit authenticationRADKIT_DEFAULT_SERVICE_SERIAL — RADKit service instance identifierRADKIT_CERT_B64 — Base64-encoded client certificateRADKIT_KEY_B64 — Base64-encoded private keyRADKIT_CA_B64 — Base64-encoded CA chainRADKIT_KEY_PASSWORD_B64 — Base64-encoded key passwordFor local use, RADKit auto-detects certificates in ~/.radkit/identities/ generated during the setup onboarding wizard (bash setup.sh in the cloned repo).
tools
Zoom meeting intelligence — correlates a live or referenced Zoom meeting discussion against NetClaw's historical meeting record (via the official Zoom Meetings MCP) and today's actual network state. Use when someone in a Zoom meeting references a past discussion or incident ('didn't we have this issue before?'), or asks to search prior meetings for a topic. Does not itself recognize live in-meeting questions — that happens automatically inside zoom-rtms-mcp's own extractor (spec 118) before this skill is ever invoked.
tools
Manage Lantronix out-of-band (OOB) infrastructure via Percepxion central management platform: device inventory, serial port inspection via SLC CLI, firmware compliance, config management, security auditing, and closed-loop incident remediation. Use during outages, maintenance windows, compliance cycles, and AI-assisted automation workflows.
tools
Federate your NetClaw with other NetClaw operators over the BGP mesh — exchange capability inventories and ask your claw what a peer can do. (US1; remote invocation and chat land in later phases.)
tools
Track token consumption, enforce session budgets, and display cost for every NetClaw interaction.