The Agent Skills Directory

[DATA_EXFILTRATION]: The skill instructs the agent to read local MCP configuration files, such as claude_desktop_config.json and .mcp.json. This access is used specifically to detect conflicts between Zapier-integrated actions and native MCP servers during the 'Audit' mode. No data is transmitted to non-vendor domains.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to manage Zapier tool states using official MCP tools like enable_zapier_action and disable_zapier_action as part of the setup optimization and troubleshooting workflows.
[PROMPT_INJECTION]: The skill contains instructional headers like 'CRITICAL' or 'IMPORTANT' in its documentation, but these are used for standard operational guidance rather than attempting to bypass model safety filters or override system constraints.

zapier-status