The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a generated command-line interface (CLI) to manage application lifecycles and health checks. Specific commands like up, down, doctor, and check are executed locally to bring the app up, run tests, and capture logs or screenshots.
[DYNAMIC_EXECUTION]: The skill guides the agent to generate a new executable script or CLI in the repository's native language. The agent then runs this dynamically created code to perform verification tasks, representing a script generation and execution pattern.
[INDIRECT_PROMPT_INJECTION]: The agent is directed to derive its verification logic by reading untrusted content within a repository, such as README files, package scripts, and other documentation. This ingestion of external data creates a potential surface for indirect instructions to influence the agent's behavior.
Ingestion points: Repository files including SKILL.md, README.md, package.json, and source code.
Boundary markers: None specified in the instructions for delimiting repo content.
Capability inventory: The agent is authorized to write files, install dependencies, and execute shell commands.
Sanitization: No specific sanitization or validation of the repository content is mentioned before it is used to generate the CLI logic.

create-verification-skill