The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's scripts and the generated test runner utilize the subprocess module to execute pytest and the claude CLI. These operations are essential for the skill's purpose of automating end-to-end tests.
[EXTERNAL_DOWNLOADS]: The generated Dockerfile includes a command to install the @anthropic-ai/claude-code package from the official NPM registry, which is a trusted and expected dependency for this skill's functionality.
[COMMAND_EXECUTION]: The generated test runner uses the --dangerously-skip-permissions flag when calling the claude CLI. This is a standard requirement for non-interactive execution in automated testing environments.
[CREDENTIALS_UNSAFE]: The generated docker-compose.yml file configures a volume mount from the host's ~/.claude directory to the container. This allows the testing environment to use existing authentication for API calls during the E2E process.

e2e-testing