The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions in SKILL.md guide the agent to run the scripts/bootstrap_slice.py Python script to perform the bootstrap operation using command-line arguments.
[REMOTE_CODE_EXECUTION]: The script scripts/bootstrap_slice.py utilizes importlib.util to dynamically load and execute Python modules (manage_execution.py and manage_planning.py) from relative filesystem paths within the vendor's skill hierarchy. This dynamic execution pattern is used to facilitate internal communication between modular skills.
[PROMPT_INJECTION]: The skill handles planning artifacts and work item descriptions, creating a surface for indirect prompt injection where instructions embedded in project documentation could influence agent behavior.
Ingestion points: The script reads planning feature metadata and directory structures from the filesystem through its sibling modules.
Boundary markers: No explicit boundary markers or 'ignore' instructions are documented for the content of the planning artifacts being processed.
Capability inventory: The skill possesses capabilities for file system modification and dynamic loading of executable Python logic.
Sanitization: Inputs such as slice IDs and feature names are passed through validation and normalization functions defined in the loaded management modules.

slice