The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill enables the agent to ingest and process data from external Miro boards, which could contain malicious instructions designed to influence agent behavior.
Ingestion points: The tools context_get, doc_get, table_list_rows, and board_list_items retrieve data from Miro boards (SKILL.md).
Boundary markers: The skill does not define or recommend using delimiters or specific 'ignore embedded instructions' warnings when processing Miro content.
Capability inventory: The skill possesses capabilities to create and update diagrams, documents, and tables on Miro boards.
Sanitization: No sanitization, escaping, or validation of the content retrieved from the boards is described in the skill.

miro-mcp