The Agent Skills Directory

[SAFE]: The skill implements normal vendor functionality for searching and reading personal documents. No malicious patterns, command injection, or unauthorized data exfiltration behaviors were found.- [INDIRECT_PROMPT_INJECTION]: The skill processes document content from the Yuque platform, which is an external data source. This is an expected behavior for a search skill.
Ingestion points: Document content is ingested via the yuque_get_doc tool as defined in the Step 4 workflow of SKILL.md.
Boundary markers: The skill does not explicitly define delimiters for retrieved content, but it specifies a rigid synthesis format for the final response.
Capability inventory: The skill's capabilities are scoped to yuque_search and yuque_get_doc within the yuque-mcp environment.
Sanitization: The instructions rely on natural language synthesis constraints without additional programmatic filtering of document content.

smart-search