The Agent Skills Directory

[SAFE]: The skill is designed for administrative and project management tasks within a workspace. It performs standard file operations (reading and writing YAML) in a dedicated directory ('planning/schedules/') and checks for directory existence in the 'openspec/changes/' folder, which is consistent with its stated purpose.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection, as it processes and displays data from external files. 1. Ingestion points: The skill parses data from 'planning/schedules/*.yaml' files during the 'review' and 'update' operations. 2. Boundary markers: There are no defined boundary markers or instructions to isolate data from the agent's executable context. 3. Capability inventory: The skill can read and write files in the local workspace and interact with the 'yunxiao' skill. 4. Sanitization: The instructions do not provide for sanitization or validation of text content found in the YAML modules before it is presented to the user or processed by the model.

workspace-planning