pyATS Network Device Tool

What it does

This skill integrates pyATS-based workflows into an agent environment to interact with network devices (Cisco IOS‑XE/NX‑OS, Linux hosts in testbeds). It exposes a small set of safe, focused tools: list devices, run parsed 'show' commands (Genie when available), fetch running configuration and logs, execute pings from devices, run Linux shell commands on capable hosts, apply controlled configuration changes, and run sandboxed dynamic pyATS tests. The emphasis is on gathering structured diagnostics and performing guarded changes with validation rules.

When to use it

Use this skill when you need automated network inventory and troubleshooting (interface status, routing, BGP/OSPF neighbors), to capture configs or logs for audits, to run device-originated pings, or when you want to apply small, validated configuration snippets. It's appropriate for CI/testbed automation, pre/post change checks, and scripted validation runs. Avoid using it for destructive ops (reload/erase) — the skill enforces blocked commands.

What's included

• Scripts: none bundled in the skill dir (has_scripts=false) but the MCP wrapper script is documented
• References: none attached (has_references=false)
• Instructions: clear tool call patterns using the MCP_CALL protocol and examples for each of eight exposed tools (list devices, run show, configure safely, show running-config, fetch logs, device ping, run Linux commands, run dynamic tests). Validation rules and banned commands are documented.

Compatible agents

Best used by automation-focused agents that can run Python and interact with network testbeds (agents with python3 and pyATS support, e.g., code-capable assistants or CLI automation agents).