The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and execute the @useclawlink/cli package from the official NPM registry to facilitate authentication and communication with the Telegram API.
[COMMAND_EXECUTION]: The skill relies on executing shell commands via npx to interact with the vendor's command-line interface for logging in, connecting services, and running Telegram actions.
[DATA_EXFILTRATION]: The skill accesses the local file path ~/.clawlink/credentials.json to store and retrieve authentication tokens used to maintain the session with the ClawLink service.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, creating a potential attack surface for indirect prompt injection.
Ingestion points: External data enters the agent context through tools like telegram_get_updates and telegram_get_chat_history, which retrieve messages from Telegram users.
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to the agent to ignore instructions embedded within the retrieved chat content.
Capability inventory: The skill has the ability to execute shell commands (via the CLI) and perform write operations to Telegram (e.g., telegram_edit_message, telegram_delete_message, telegram_create_chat_invite_link).
Sanitization: There is no mention of sanitization or filtering of the content retrieved from Telegram before it is processed by the agent.

telegram