The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface due to its ability to ingest and process external document data.\n
Ingestion points: Document and Wiki content is retrieved through the docs +fetch shortcut as documented in references/lark-doc-fetch.md.\n
Boundary markers: There are no explicit instructions for using delimiters or boundary markers to separate retrieved document content from system prompts or instructions.\n
Capability inventory: The skill allows for significant modifications and creation of resources via lark-cli commands like +create, +update, and +media-insert, as detailed in references/lark-doc-create.md, references/lark-doc-update.md, and references/lark-doc-media-insert.md.\n
Sanitization: The provided instructions do not include mechanisms for sanitizing or validating external content before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill relies on the lark-cli tool, which is a required binary specified in the skill metadata. This tool is used to execute all document management tasks, including file system access for media uploads as seen in references/lark-doc-media-insert.md and references/lark-doc-media-download.md.

lark-doc