The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill possesses a surface for indirect prompt injection.
Ingestion points: Task data and project status are ingested via get_tasks and get_project_status as defined in references/ohno-tools.md.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands in the ingested data are present.
Capability inventory: The skill can modify the project state using tools like update_task, archive_task, and add_dependency.
Sanitization: No sanitization or validation of the input data is specified.
Mitigation: The workflow includes mandatory 'Execute with approval' checkpoints which reduces the risk of autonomous malicious actions.

plan-revision