The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing shell commands to interact with Git and run Node.js-based linting tools.\n
Evidence: git diff --cached --name-only is used to generate file lists.\n
Evidence: npx eslint, npx prettier, and npx stylelint are executed against these file lists.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes the content of code files during the linting process. If these files contain malicious payloads (e.g., in comments), they could influence the agent when it reviews the linter's output.\n
Ingestion points: Changed files identified in Step 1 of the skill workflow.\n
Boundary markers: The skill instructions specify presenting findings in Markdown tables, which helps but does not eliminate injection risks.\n
Capability inventory: The skill has the ability to write to files (via fix flags) and execute shell commands.\n
Sanitization: There is no explicit logic described to sanitize the output of the linting tools before the agent processes it.\n- [PERSISTENCE]: The skill suggests configuring Git hooks, which results in automated code execution every time a user performs a commit.\n
Evidence: Recommendation to use npx husky add .husky/pre-commit to automate checks.

enforcing-code-linting