The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external content from Telegram users, which could contain instructions intended to manipulate the agent.\n
Ingestion points: The skill utilizes TELEGRAM_GET_UPDATES and TELEGRAM_GET_CHAT_HISTORY to read messages from external chats.\n
Boundary markers: There are no instructions for using delimiters or specific warnings to ignore instructions embedded in the chat data.\n
Capability inventory: The agent is granted capabilities to send, edit, and delete messages, as well as manage bot commands through tools like TELEGRAM_SEND_MESSAGE and TELEGRAM_SET_MY_COMMANDS.\n
Sanitization: No security-focused sanitization or filtering of incoming data is mentioned to prevent instruction leakage into the agent prompt.\n- [EXTERNAL_DOWNLOADS]: The skill relies on an external service endpoint for its core functionality.\n
The setup requires adding https://rube.app/mcp as an MCP server to the agent configuration.

telegram-automation