Can I see a preview of the commands before they are executed?

Yes. Every invocation generates a clear Execution Plan that categorizes commands with status indicators, showing you exactly what will run, what requires confirmation, and what will be skipped.

How does Command Executor ensure my system stays safe?

The skill utilizes a strict safety policy that blocks dangerous commands like 'rm -rf' or 'drop database' and requires explicit user confirmation for any operation that modifies files or installs dependencies.

Where are the execution logs and audit trails stored?

All execution data, including output captures and timestamps, are stored locally in the '.cache/command-validations/' directory for easy auditing and offline access.

Does this skill work with non-JavaScript projects?

While it includes specific optimizations for Node.js workflows, it is a cross-platform tool capable of discovering and executing any standard shell commands found in your project's markdown documentation.

How does the performance regression detection work?

The skill saves execution times to a local cache and compares the current run's duration against previous benchmarks. If a command takes more than 1.5x its usual time, it flags it as a potential regression.

Command Executor

Name: Command Executor
Author: d-oit

byd-oit

•

セキュリティとテスト

Executes and validates terminal commands safely while capturing real output and monitoring for performance regressions.

Command Executor is a specialized Claude Code skill designed to bridge the gap between identifying shell commands and running them with confidence. It implements a robust safety layer by categorizing commands into safe, conditional, or dangerous groups, effectively preventing accidental data loss or destructive operations. The skill automates the execution of build scripts, tests, and linters while capturing granular metrics—including stdout, stderr, exit codes, and duration—into a persistent local cache. By integrating with git-aware change detection, it ensures that your project documentation remains accurate and your scripts remain performant across different environments.

主な機能

01Performance regression detection via execution duration benchmarking

021 GitHub stars

03Comprehensive output capture and audit logging for all terminal operations

04Intelligent command categorization into safe, conditional, and dangerous tiers

05Automated execution planning with mandatory user confirmation for risky tasks

06Git-aware pre-flight checks to prevent accidental execution on production branches

ユースケース

01Automating repetitive build and test cycles with built-in performance monitoring

02Verifying that code examples in documentation are functional and up-to-date

03Executing complex CLI workflows with safety rails against destructive operations

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add d-oit/command-verify command-executor

For use in Claude.ai and ChatGPT

Download Skill