The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data to generate test plans, creating an attack surface where malicious instructions in documentation could influence agent behavior.
Ingestion points: The skill loads PRD.md, Epics.md, Stories.md, architecture.md, and tech-spec documents (found in steps-c/step-02-load-context.md).
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands when these documents are read into the agent's context.
Capability inventory: The skill has access to read_file, write_file, list_files, search_repo, and shell execution tools like uv run and playwright-cli.
Sanitization: There is no evidence of sanitization or content filtering for the ingested documents.
[DYNAMIC_EXECUTION]: The skill uses computed paths to execute Python scripts at runtime for configuration management.
Evidence: Instructions in SKILL.md direct the agent to execute uv run {project-root}/_bmad/scripts/resolve_customization.py and resolve_config.py.
[EXTERNAL_DOWNLOADS]: The skill references an external package for migration tasks.
Evidence: SKILL.md instructs the user or agent to run npx --package=bmad-method-test-architecture-enterprise@3 tea-migrate-customizations. This package belongs to the vendor's organization.
[COMMAND_EXECUTION]: The skill frequently uses shell-based tooling for automation and project setup.
Evidence: The workflow involves executing uv run, npx, and playwright-cli commands to perform its primary functions.

bmad-tea-test-design