pktctl

Overview Schema Source Related Servers Discussions Admin

run_cli

Destructive

Runs one IOS command on a Packet Tracer device console, returning output and status. Handles long-running commands and interactive prompts.

Instructions

Type one IOS command at the console of a router or switch and return its output, including commands that take time such as ping or traceroute. status tells whether IOS accepted the command. A command still running after timeout_secs is stopped with Ctrl+Shift+6 and comes back with finished: false and the output so far. When IOS asks something ([confirm], [yes/no], Password:) the reply carries question and the console keeps waiting: answer with another run_cli in mode current, an empty command for Enter.

Input Schema

TableJSON Schema

Name	Required	Description
`mode`	No	Mode to run the command in. Defaults to `enable`.
`device`	Yes	Device name exactly as shown by `list_devices`.
`command`	Yes	One IOS command, for example `show ip interface brief`. To answer a question the previous command left open (`question` in its reply), send the answer here with mode `current`; an empty command presses Enter, which confirms `[confirm]`.
`password`	No	Console line password, when the device asks for one at `Password:`.
`timeout_secs`	No	Seconds to wait for the command to finish. Defaults to 30, maximum 300.
`enable_password`	No	Privileged mode password (`enable secret` or `enable password`).

Output Schema

TableJSON Schema

Name	Required	Description
`output`	Yes
`status`	No
`finished`	Yes
`question`	No	The question the command is waiting on, such as `Proceed with reload? [confirm]`. Answer it with another call in `current` mode; an empty command presses Enter.

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.2.2Sep 28, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the annotations: it discloses timeout handling (Ctrl+Shift+6 abort, `finished: false` with partial output), the `status` acceptance field, and the multi-turn question/answer protocol. This is exactly the kind of behavior an agent could not infer from readOnlyHint/destructiveHint.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action, then the timeout and question-handling behavior in dense, purposeful sentences. Every sentence carries information, though the interactive-prompt paragraph is slightly compressed and takes a second read.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With an output schema present, return values need not be spelled out, yet the description still names the key reply fields (status, finished, question) and covers the tricky interactive loop. Nothing needed to call this correctly appears to be missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% so the baseline is 3, but the description adds cross-parameter semantics the schema does not: how `command` and mode `current` combine to answer a pending question, and that an empty command sends Enter to confirm a `[confirm]` prompt. That said, it does not clarify the user/enable/global mode distinctions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Type one IOS command at the console of a router or switch and return its output') and scopes it to IOS devices, which cleanly separates it from run_host_command and the configuration-oriented siblings like configure_ios.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explains the context of use (single commands, including slow ones like ping/traceroute) and gives an explicit when-to-use rule for the interactive path: answer a pending `question` with another run_cli in mode `current`. It does not name run_host_command as the alternative for non-IOS hosts, so routing between the two is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server