Skip to content

Terminal Capture

Automates screenshots of CLI and terminal UI output for visual checks and PR reviews

Stars
28,880
Licence
Apache-2.0
Files
1 file

What it does

When installed, the agent can drive the qwen-code command-line tool through scripted scenarios: it types text and keys into a pseudo-terminal, waits for output, and saves screenshots (and optionally an animated GIF) of the terminal. This lets the agent visually verify how CLI commands such as /about or /export look, which is useful when reviewing pull requests that change terminal output. The skill is set up for the qwen-code repository and relies on its TypeScript test scripts.

Great for

  • Maintainers of qwen-code reviewing pull requests that change CLI or slash-command output
  • Developers who want screenshots of terminal interfaces for documentation
  • Teams that need repeatable visual checks of interactive terminal flows

Skip it if

  • Someone without the qwen-code repository checked out, since the scripts and paths it references live there
  • Running with real API keys in shared or CI environments, because scenarios can type credentials such as sk-xxx into the terminal and save them in screenshots
  • Simple text-only checks where reading the command output directly is enough

Try asking

  • “Take screenshots of the /about command for this PR and show me the results”
  • “Run the terminal capture scenarios in integration-tests/terminal-capture/scenarios and summarise any visual changes”
  • “Record an animated GIF of the progress bar output from my bash progress script”
  • “Check how the /export format selection menu looks after my change”
Step-by-step guide: how to use Terminal CaptureInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    terminal-capture.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip terminal-capture.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip terminal-capture.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Review scenario files before running, since they execute the spawn command and any typed input, and avoid putting real API keys in scenario files or screenshots
  • In CI verify runs, rely on the pre-installed browser and do not run playwright install, as the skill itself advises

What's inside

  • SKILL.md

Redistributed unmodified from QwenLM/qwen-code under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.