The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing shell commands via the mcporter utility to interface with the SiYuan Note API. These commands are used for standard note-taking operations such as searching (unified_search), reading (get_document_content), and appending text (append_to_daily_note).
[EXTERNAL_DOWNLOADS]: The documentation links to the author's GitHub repository (https://github.com/PurpleLiu/siyuan-mcp) to provide the required MCP server. As this repository belongs to the skill's author, it is considered a legitimate vendor resource.
[PROMPT_INJECTION]: The workflows for organizing notes and generating weekly reports involve reading existing note content to produce summaries. This creates an indirect prompt injection surface where malicious instructions inside a note could potentially influence agent behavior.
Ingestion points: get_document_content is used in Workflows E and F.
Boundary markers: The skill does not implement delimiters or safety warnings for the LLM when processing retrieved note content.
Capability inventory: The agent has the ability to read, create, and write to the local note-taking database via shell commands.
Sanitization: No explicit content filtering or sanitization is performed on the notes before they are processed by the LLM.

siyuan-skill