inbox-rotation-scaling
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and fetches metadata from
manypi.comandmcp.manypi.com. These domains are owned by the skill's author (manypicom). - [REMOTE_CODE_EXECUTION]: The skill provides a command to add an MCP server from
https://mcp.manypi.com/mcp. This pattern involves loading external tool definitions and logic from a remote URL to extend the agent's capabilities. - [COMMAND_EXECUTION]: Instructs the user to run
claude mcp add, which modifies the agent's runtime environment to include new tools for mailbox management. - [INDIRECT_PROMPT_INJECTION]: The skill involves managing email sequences and processing replies. Because the agent processes incoming content from external recipients, it represents an attack surface where third-party data could attempt to influence the agent's instructions.
Audit Metadata