The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface by reading untrusted content from external sources.
Ingestion points: Untrusted data enters the agent's context through tools such as TELEGRAM_GET_UPDATES and TELEGRAM_GET_CHAT_HISTORY (referenced in SKILL.md).
Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions embedded within the retrieved Telegram messages.
Capability inventory: The skill provides significant execution capabilities across several tools, including TELEGRAM_SEND_MESSAGE, TELEGRAM_SEND_PHOTO, TELEGRAM_SEND_DOCUMENT, TELEGRAM_DELETE_MESSAGE, TELEGRAM_SET_MY_COMMANDS, and TELEGRAM_EXPORT_CHAT_INVITE_LINK.
Sanitization: There is no evidence of sanitization, validation, or filtering of the content retrieved from Telegram before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill requires the configuration of an external MCP server located at https://rube.app/mcp. This represents a dependency on an external endpoint that is not within the whitelisted or trusted domains.
[COMMAND_EXECUTION]: The skill facilitates the execution of various Telegram operations that can modify chat states or broadcast information, such as deleting messages, managing chat admins, and setting bot commands.

telegram-automation