The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes recently modified code and applies refinements to it. This creates a surface for indirect prompt injection, as malicious instructions embedded within code comments or string literals could potentially influence the agent's behavior. The instructions do not specify boundary markers or sanitization procedures for the processed data.
[COMMAND_EXECUTION]: The instructions direct the agent to 'operate autonomously and proactively,' refining code immediately after it is written without requiring explicit requests. This behavior reduces user oversight and could lead to unintended or unreviewed file modifications, although the scope is limited to code refinement.

simplify