The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides structured instructions for interacting with the Feishu/Lark environment through the lark-cli binary. It includes numerous examples of shell commands for meeting search (vc +search), note retrieval (vc +notes), and recording management (vc +recording).
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external and untrusted meeting data, including transcripts, AI summaries, and shared documents, which presents a surface for indirect prompt injection.
Ingestion points: External content is ingested via lark-cli docs +fetch and lark-cli vc +notes as described in SKILL.md and references/lark-vc-notes.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within the meeting content.
Capability inventory: The skill uses tools with capabilities to read/write files (e.g., media-download for whiteboard covers) and perform network API calls via the CLI tool.
Sanitization: There are no explicit sanitization or filtering steps mentioned for the retrieved meeting content before it is processed by the agent.

lark-vc