The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by reading and processing the content of GitHub Actions workflow files during the linting process.
Ingestion points: Content of staged files located in .github/workflows/ (SKILL.md).
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the ingested workflow files.
Capability inventory: Executes shell commands via git and the actionlint binary (SKILL.md).
Sanitization: No explicit sanitization or filtering is performed on the workflow content before it is passed to the linter or reported to the agent.
[COMMAND_EXECUTION]: Executes shell commands including git diff, git rev-parse, and actionlint to identify staged changes and perform syntax validation.
[EXTERNAL_DOWNLOADS]: References and suggests the installation of the actionlint binary and the pre-commit framework. It correctly directs the agent to use the mise tool manager and official GitHub sources, explicitly advising against insecure installation practices like curl | bash.

actionlint