control-cli

An agent skill by cursor, from cursor/plugins. Tags: automation, cli, debugging, testing, validation.

What it does

Build or adapt a local harness to drive, inspect, and profile an interactive CLI or TUI without external services. Use for CLI UX checks, startup regressions, memory leaks, hangs, prompt flows, or terminal demos.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add cursor/plugins --skill control-cli

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/cursor/plugins
cp -r plugins/cursor-team-kit/skills/control-cli ~/.claude/skills/control-cli

Safety box score

Grade B, low risk: 11 of 100 risk points, rated 2026-10-19. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill sets up local tmux/PTY harnesses to run an interactive CLI, send keystrokes, and capture output or profiles. The main risk is local command execution against whatever CLI you point it at; it does not describe network access, privilege escalation, or data leaving the machine.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It explicitly runs target commands in `tmux new-session -d -s "$SESSION" -- <command-under-test>` and via a Python `subprocess.Popen([...])` PTY harness, so it can execute local programs and scripts.
Network access	0 (none)	The skill says to work locally and mentions no downloads or remote hosts; the only shown commands are `tmux`, PTY, and local profiling.
Filesystem reach	1 (low)	It saves transcripts/profile artifacts and says to keep the harness in `/tmp`, but does not instruct writes outside the working area or system paths.
Secrets & credentials	0 (none)	It explicitly says, "Do not send credentials," and there is no step that reads or requests secrets, keychains, or login stores.
Privilege & persistence	0 (none)	There is no sudo, service installation, shell startup editing, or system configuration change in the skill.
Instruction hijack surface	0 (none)	No hidden instructions, instruction-skipping, or fetched remote prompts appear in the skill text.
Irreversible actions	0 (none)	The skill only drives and inspects a local CLI/TUI; it does not push, post, deploy, pay, book, or delete anything.
Transparency	0 (none)	All included material is plain text; the tmux and PTY examples are readable and there are no binaries or decoded blobs.

Before installing

Check whether the target CLI command itself is safe to run unattended, since this skill will execute it in a live terminal session.
Confirm any transcripts/profile outputs will stay in a scratch location such as `/tmp` and not in a synced or sensitive directory.
If you plan to profile a Node CLI, verify the `--inspect` port exposure is acceptable on your machine.
Read the repo’s own harness scripts first, because the skill prefers reusing them over a temporary wrapper.

Rated 2026-10-19 by an automated review of 1 file from the skill's repository.

Source

Repository: cursor/plugins (all skills from this repository)
Path: cursor-team-kit/skills/control-cli/SKILL.md
Branch: main
Collection: cursor
Updated: 2026-09-23

E2E Testing Nori Skillsets Subcommands — Use when you need to interactively test a nori-skillsets CLI subcommand end-to-end via tmux, with full filesystem isolation
TUI Puppeteering with tmux — Use when automating or testing TUI/CLI applications - provides isolated tmux sessions with scripts for input, output capture, and state verification
check-compiler-errors — Run compile and type-check commands and report failures
control-ui — Build or adapt a local browser/CDP harness to drive and inspect a web, IDE, or Electron UI.
reproduce-and-fix-issues — Reproduce triaged Slack bugs through a configured app-control adapter, verify existing fixes.
vitest — Vitest fast unit testing framework powered by Vite with Jest-compatible API.