The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts, specifically resolve_customization.py and resolve_config.py, located within the project's _bmad/scripts/ directory using the uv run command. These are internal utility scripts used to manage project-specific configurations.
[EXTERNAL_DOWNLOADS]: The workflow facilitates the installation of standard development dependencies and tools. This includes Node.js packages like @seontechnologies/playwright-utils and @seontechnologies/pactjs-utils, as well as standard frameworks such as Playwright and Cypress. It also provides templates for downloading tools like Maestro via curl in CI environments.
[DYNAMIC_EXECUTION]: The skill installs a write-time enforcement hook, tea-enforce.cjs, into the project's .claude/hooks/ directory. This script is registered in .claude/settings.json to monitor and validate test code as it is being written, preventing anti-patterns like hard waits or tautological assertions.
[INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect injection as it reads and parses various project manifest files (e.g., package.json, pyproject.toml, go.mod) to auto-detect the project stack. However, this is a necessary function for its stated purpose, and the skill includes validation checks for the generated outputs.

bmad-tea-framework