skills/openhands/skills/discord/Gen Agent Trust Hub

discord

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill communicates with the official Discord API (discord.com). It implements specific safety measures to redact sensitive webhook URLs—which contain authentication tokens—from exception messages and logs to prevent credential leakage.
  • [COMMAND_EXECUTION]: Provides standard CLI examples using curl and bundled Python scripts to post messages. These scripts are self-contained and focus on the primary purpose of Discord integration.
  • [EXTERNAL_DOWNLOADS]: The bundled Python scripts utilize the requests library for network communication. All network activity is directed toward legitimate Discord infrastructure.
  • [PROMPT_INJECTION]: The skill handles message content which may originate from untrusted sources. It addresses potential risks (such as unauthorized pings) by defaulting the allowed_mentions parameter to an empty list, preventing the agent from accidentally triggering @everyone or role mentions unless explicitly configured.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 01:31 PM