The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reads and processes all documentation, source code, and test files within a project directory, creating a surface for indirect prompt injection.
Ingestion points: The skill is instructed to read all stage documents, project code, and test files as listed in the 'Input' section of SKILL.md.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions that might be embedded within the project files (e.g., in markdown comments or code strings).
Capability inventory: The skill uses Bash(*) to execute tests and linting tools, providing a significant execution capability if the agent is misled by injected content.
Sanitization: No sanitization or verification of the external file content is performed before the agent uses it to determine execution steps.
[COMMAND_EXECUTION]: The skill leverages Bash(*) to perform its core functions, including running unit tests, integration tests, and linting tools. This broad permission is necessary for the skill's primary purpose but represents a powerful capability that requires careful oversight.

self-verification