The Agent Skills Directory

Unverifiable Dependencies (MEDIUM): The skill requires the user to add an external MCP server endpoint https://rube.app/mcp. This domain is not included in the list of trusted providers (e.g., Anthropic, Google, Microsoft, Composio) and acts as a remote dependency that processes all tool calls and data. While the documentation claims no API keys are needed, routing sensitive presentation data through an unverified relay poses a supply-chain risk.
Indirect Prompt Injection (LOW): The skill possesses a significant attack surface for indirect prompt injection.
Ingestion points: The markdown_text parameter in GOOGLESLIDES_CREATE_SLIDES_MARKDOWN and GOOGLESLIDES_PRESENTATIONS_BATCH_UPDATE tools accepts arbitrary text that the agent is instructed to process.
Boundary markers: The skill does not define delimiters or provide 'ignore embedded instructions' warnings for the data being ingested.
Capability inventory: The skill has broad capabilities to create, modify, retrieve, and share Google Slides presentations and Google Drive files.
Sanitization: There is no evidence of sanitization or validation of the input Markdown before it is passed to the underlying Google Slides API tools.
Risk: An attacker could embed malicious instructions in a Markdown source (e.g., a document the user asks the agent to summarize into a presentation) that could cause the agent to exfiltrate slide content or share presentations with unauthorized users.

googleslides-automation