The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by ingesting data from project files to generate new source code.\n
Ingestion points: The skill reads untrusted data from tests/**/*_test.*, design/gdd/systems-index.md, docs/architecture/tr-registry.yaml, and various system-specific GDD files in SKILL.md.\n
Boundary markers: The instructions do not specify any boundary markers or delimiters to prevent the agent from obeying instructions that might be embedded within the project files it reads.\n
Capability inventory: The skill is granted Write permissions to create new files within the project's tests/helpers/ directory.\n
Sanitization: There is no evidence of sanitization, validation, or filtering of the content extracted from external project files before it is interpolated into the generated code templates.\n- [SAFE]: The skill implements protective measures for filesystem operations by explicitly stating it will never overwrite existing files and requiring user confirmation before performing any write actions.

test-helpers