discord
Pass
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from an external source.
- Ingestion points: The
scripts/messages.rbscript retrieves message content from Discord channels and presents it to the agent. - Boundary markers: Absent. External message content is not wrapped in delimiters or accompanied by instructions to ignore embedded commands.
- Capability inventory: The skill possesses several sensitive capabilities, including sending (
send_message.rb), editing (edit_message.rb), and deleting (delete_message.rb) messages, as well as creating threads (thread_create.rb) and managing reactions (react.rb). - Sanitization: The scripts do not perform sanitization or validation of the retrieved message content.
- [SAFE]: The skill implements secure credential management by storing the Discord bot token in a local directory (
~/.config/discord/token) with restricted file permissions (0600), ensuring the token is only accessible by the current user. - [SAFE]: Network activity is exclusively directed to the official Discord API domain (
discord.com), which is an expected and well-known service for the skill's stated purpose.
Audit Metadata