The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several local utility scripts (detect-project.sh, generate-install-cmd.sh, categorize-errors.js, validate-setup.sh) to analyze the project structure, generate installation commands, and prioritize linting errors. These scripts use standard Node.js and Bash patterns to process project metadata.
[EXTERNAL_DOWNLOADS]: The skill triggers the installation of standard, well-known development dependencies from the official npm registry. These include eslint, typescript-eslint, and various community-standard plugins (e.g., eslint-plugin-react, eslint-plugin-unicorn, eslint-plugin-security) based on the detected project environment.
[COMMAND_EXECUTION]: The skill supports advanced integration with the Claude Code platform by installing a local hook script (lint-typecheck-hook.sh) and configuring .claude/settings.json. This hook facilitates real-time linting of edited files, using a 'circuit breaker' pattern in a temporary directory to prevent infinite loops.
[COMMAND_EXECUTION]: Verification steps involve running standard project commands like npm run lint and npx tsc --noEmit to ensure the project remains in a valid state after automated fixes.
[DATA_EXFILTRATION]: No evidence of unauthorized network calls or sensitive data access was found. The skill focuses exclusively on project-local configuration and code quality improvements.

ts-linter