The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted message content from the Telegram API which could contain malicious instructions designed to manipulate the agent.
Ingestion points: The script scripts/fetch_dm.py retrieves message text from the Telegram getUpdates API endpoint, which can contain any text sent to the monitored channel.
Boundary markers: Absent. The message content is printed directly to stdout and incorporated into the agent's context without delimiters or instructions to ignore embedded commands.
Capability inventory: The skill provides scripts/send_channel.py for posting to channels and likely operates within an environment where the agent has broader file system or network access.
Sanitization: Minimal. The script only uses a regex to strip leading bot mentions, providing no protection against instructions embedded in the rest of the message body.

telegram-bot