The Agent Skills Directory

[PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface due to its core functionality of reading external file contents.
Ingestion points: The agent ingests untrusted data through the DROPBOX_READ_FILE tool and file search metadata.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions embedded within the Dropbox files it processes.
Capability inventory: The skill has access to impactful tools including DROPBOX_DELETE_FILE_OR_FOLDER, DROPBOX_UPLOAD_FILE, and DROPBOX_CREATE_SHARED_LINK across all documented workflows.
Sanitization: No sanitization or validation logic is specified for data retrieved from the remote storage before it is added to the agent's context.
[SAFE]: The skill describes a standard integration with a third-party MCP server (Rube MCP) to provide cloud storage automation. All external references are directly related to the skill's stated purpose of Dropbox automation.

dropbox-automation