The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes meeting transcripts and notes which are external, potentially untrusted data sources. Malicious content within a meeting's verbatim transcript could attempt to influence the agent's summarization or document creation behavior.
Ingestion points: Meeting notes and transcripts are retrieved via lark-cli note +detail and lark-cli minutes +detail.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the retrieved text are defined in the workflow.
Capability inventory: The agent has the ability to write to the local filesystem (--output-dir ./transcripts) and create or update cloud documents (lark-cli docs +create/update).
Sanitization: The instructions do not specify any sanitization or filtering of the meeting content before it is interpolated into the final report or document content.
[COMMAND_EXECUTION]: The skill relies on executing the lark-cli tool and system commands like date to perform its tasks. These are documented as standard operations for the skill's functionality.

lark-workflow-meeting-summary