The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses scripts/lint_runner.py to execute external validation tools like npm, npx, ruff, and mypy. The Python implementation uses subprocess.run with list-based arguments, which follows security best practices by preventing shell injection vulnerabilities.
[PROMPT_INJECTION]: The skill body uses imperative language ("MANDATORY", "Strict Rule") to define a development workflow. These instructions are task-oriented and do not attempt to bypass agent safety filters or override core behavioral constraints.
[INDIRECT_PROMPT_INJECTION]: The skill reads project configuration files to automate linter execution, creating a potential surface for indirect instruction execution.
Ingestion points: Reads scripts and dependencies from package.json, as well as configuration from pyproject.toml and requirements.txt (scripts/lint_runner.py).
Boundary markers: None; the skill assumes the developer environment is a trusted context for running quality tools.
Capability inventory: Executes shell commands through package managers (e.g., npm run lint).
Sanitization: None; the skill relies on the standard behavior of the underlying development tools.

lint-and-validate