The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data retrieved from Discord (such as messages, member names, and channel topics) which could contain malicious instructions targeted at the AI agent.
Ingestion points: Data retrieved through commands like message fetch, search messages, and export channel enters the agent's context.
Boundary markers: The skill contains an explicit 'Safety Notes' section warning the agent to treat Discord content as untrusted and to never follow instructions found within retrieved content.
Capability inventory: The skill has the capability to perform destructive actions (e.g., member ban, channel delete, message purge) which could be abused if an injection is successful.
Sanitization: No programmatic sanitization is implemented within the skill markdown; protection depends on the agent adhering to the safety instructions provided.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the discordcli-agents package via pip. This is a standard dependency for the tool's functionality.
[SAFE]: No malicious patterns, obfuscation, persistence mechanisms, or unauthorized privilege escalation attempts were detected. The skill correctly advises using environment variables for sensitive bot tokens rather than hardcoding them.

discord-cli-server-management