The Agent Skills Directory

[SAFE]: The skill is a set of guidelines for task management and context retention. No malicious behavior or high-risk commands were identified within the instructions.
[NO_CODE]: The skill does not include any executable scripts, binaries, or external package dependencies (npm, pip, etc.).
[PROMPT_INJECTION]: The skill creates a surface for both direct and indirect prompt injection by instructing the agent to treat a tracking document ($path) and user arguments as the 'source of truth'.
Ingestion points: Content read from the tracking document at $path and instructions provided in the argument-hint field.
Boundary markers: The skill lacks explicit markers or delimiters to differentiate between task data and potential embedded instructions.
Capability inventory: The agent is authorized to create/update files and monitor local Git repository history.
Sanitization: No input validation or instruction-filtering is applied to the tracking document's content.

working-on