The Agent Skills Directory

[SAFE]: No security vulnerabilities or malicious patterns were identified. The skill's scope is strictly limited to reading and modifying local planning files within the user's workspace.
[PROMPT_INJECTION]: The skill uses directive language such as "non-negotiable" and "overrides plan-only rules" to guide the agent's behavior for task-specific prioritization. These instructions are focused on ensuring the agent modifies the correct files during the planning workflow and do not attempt to bypass core safety or ethical constraints.
[DATA_EXFILTRATION]: There are no network-related commands (e.g., curl, wget) or external endpoints. The skill does not access sensitive system files or environment variables.

iterate-sketch