How do I install the Terminal Execution skill?

Run npx skillmds@latest add majiayu000/terminal-execution in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Terminal Execution safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Terminal Execution?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Terminal Execution free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Terminal Execution?

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Terminal Execution by majiayu000 — Agent Skill

Name: Terminal Execution
Rating: 4.2 (5 reviews)
Author: majiayu000

Terminal Execution

Best practices for reliable terminal command execution and output capture. Use this skill when running shell commands, especially in environments like WSL where output might be truncated or lost, to ensure results are properly captured and inspected.

majiayu000 Updated Sep 9, 2026 567 repo stars

File contents

Terminal Execution Guidelines

This skill defines the standard workflows for executing terminal commands to ensure reliability and visibility of outputs, particularly in complex environments like WSL.

1. Primary Execution Method

Always prioritize structured execution tools over raw terminal input when possible.

Use execute_command: Call the execute_command tool from the terminal-runner MCP server. This provides the most reliable way to get exit codes and standard output programmatically.
Fallback: If terminal-runner is unavailable, use the standard run_in_terminal tool.

2. Handling Missing Output (WSL/Complex Environments)

In some environments (like WSL), terminal output may not be returned to the agent context correctly, or it may be truncated. If you suspect output is missing:

File-Based Capture Workflow

Prepare Directory: Ensure .tmp/copilot/ exists.
Redirect Output: Run the command redirecting both stdout and stderr to a file.
Read Result: Use file reading tools to inspect the output.

Common Patterns

Capture stdout and stderr:

some-command > .tmp/copilot/output.txt 2>&1

Capture and view live (if interactive):

some-command 2>&1 | tee .tmp/copilot/output.txt

Capture exit code separately:

some-command > .tmp/copilot/output.txt 2>&1; echo $? > .tmp/copilot/exit-code.txt

3. Best Practices

File Naming: Use descriptive names for capture files (e.g., .tmp/copilot/build-log.txt, .tmp/copilot/test-results.txt).
Cleanup: Periodically clean up the .tmp/copilot/ directory to avoid confusion with stale data.

---name: terminal-executiondescription: Best practices for reliable terminal command execution and output capture. Use this skill when running shell commands, especially in environments like WSL where output might be truncated or lost, to ensure results are properly captured and inspected.--- # Terminal Execution Guidelines This skill defines the standard workflows for executing terminal commands to ensure reliability and visibility of outputs, particularly in complex environments like WSL. ## 1. Primary Execution Method Always prioritize structured execution tools over raw terminal input when possible. 1. **Use `execute_command`**: Call the `execute_command` tool from the `terminal-runner` MCP server. This provides the most reliable way to get exit codes and standard output programmatically. 2. **Fallback**: If `terminal-runner` is unavailable, use the standard `run_in_terminal` tool. ## 2. Handling Missing Output (WSL/Complex Environments) In some environments (like WSL), terminal output may not be returned to the agent context correctly, or it may be truncated. If you suspect output is missing: ### File-Based Capture Workflow 1. **Prepare Directory**: Ensure `.tmp/copilot/` exists. 2. **Redirect Output**: Run the command redirecting both stdout and stderr to a file. 3. **Read Result**: Use file reading tools to inspect the output. ### Common Patterns **Capture stdout and stderr:** ```bash some-command > .tmp/copilot/output.txt 2>&1 ``` **Capture and view live (if interactive):** ```bash some-command 2>&1 | tee .tmp/copilot/output.txt ``` **Capture exit code separately:** ```bash some-command > .tmp/copilot/output.txt 2>&1; echo $? > .tmp/copilot/exit-code.txt ``` ## 3. Best Practices - **File Naming**: Use descriptive names for capture files (e.g., `.tmp/copilot/build-log.txt`, `.tmp/copilot/test-results.txt`). - **Cleanup**: Periodically clean up the `.tmp/copilot/` directory to avoid confusion with stale data.

Frequently asked questions