The Agent Skills Directory

[SAFE]: The skill operates entirely within the Yuque ecosystem using the yuque-mcp server tools. It reads document metadata (created and updated dates) and writes a generated report to a user-specified repository.
[INDIRECT_PROMPT_INJECTION]: The skill processes document titles and knowledge base names retrieved from Yuque to populate its report. While this constitutes processing external data, the risk is minimal as the data is used to fill a structured Markdown table without high-privilege downstream execution.
Ingestion points: Document titles and repo names via yuque_list_docs and yuque_list_repos (SKILL.md).
Boundary markers: None explicitly defined in the template for the content being summarized.
Capability inventory: Ability to create new documents via yuque_create_doc.
Sanitization: No specific sanitization or escaping of document titles is mentioned before interpolation into the report body.
[CREDENTIALS_SAFE]: The skill mentions the requirement for a personal token to authenticate the yuque-mcp server in its metadata, but it follows best practices by not hardcoding any secrets or requesting them from the user during runtime.

personal-weekly