The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires connecting to an external MCP server at https://rube.app/mcp. This endpoint provides the tools necessary for Miro automation but introduces a remote dependency.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it reads untrusted content from Miro boards.
Ingestion points: MIRO_GET_BOARD_ITEMS, MIRO_GET_BOARDS2, and MIRO_GET_BOARD are used to retrieve potentially attacker-controlled text from board names and item content.
Boundary markers: No delimiters or safety instructions are defined to separate external board data from the agent's primary instructions.
Capability inventory: The skill possesses the ability to share boards via MIRO_SHARE_BOARD and modify items, which could be exploited through malicious content in board data.
Sanitization: There is no evidence of filtering or sanitizing the retrieved Miro content before it is processed by the agent.
[DATA_EXFILTRATION]: The skill facilitates the handling of sensitive data, including board contents and user emails. The MIRO_SHARE_BOARD capability could be misused to exfiltrate access to private boards if the agent is influenced by embedded instructions in the data it reads.

miro-automation