The Agent Skills Directory

[SAFE]: The skill instructs the agent to perform Zoom automation tasks through the Rube MCP framework. No malicious patterns were detected in the instructions or tool sequences.
[EXTERNAL_DOWNLOADS]: The skill setup requires adding a third-party MCP server hosted at https://rube.app/mcp. This external resource is the primary provider of the Zoom tools described and is necessary for the skill's functionality.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from Zoom (e.g., meeting topics, participant names, and recording metadata) which could potentially contain embedded instructions. The skill mitigates this risk by explicitly instructing the agent to confirm meeting details and recording consent before performing any write or delete operations.
Ingestion points: Data retrieved via ZOOM_LIST_MEETINGS, ZOOM_GET_PAST_MEETING_PARTICIPANTS, and ZOOM_LIST_ALL_RECORDINGS (in SKILL.md).
Boundary markers: The skill provides high-level instructions to verify state and intent before writes, but does not define technical delimiters for external data.
Capability inventory: The agent has the ability to create, update, and delete meetings and recordings using ZOOM_CREATE_A_MEETING, ZOOM_UPDATE_A_MEETING, and ZOOM_DELETE_MEETING_RECORDINGS (in SKILL.md).
Sanitization: No technical sanitization or filtering of meeting content is specified prior to processing.

zoom-automation