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

discord

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides Python scripts and shell recipes to interact with the Discord API. These scripts use environment variables for authentication and perform network operations to send messages.
  • [EXTERNAL_DOWNLOADS]: The bundled Python scripts depend on the requests library for HTTP communication. All network requests target official Discord API endpoints (discord.com), which is a well-known and expected service for this skill's functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data to generate message content for Discord.
  • Ingestion points: Message content is ingested from the --content CLI argument or sys.stdin in post_webhook.py and send_message.py.
  • Boundary markers: The content is interpolated directly into a JSON payload without specific delimiters.
  • Capability inventory: The skill can perform network POST requests to Discord servers.
  • Sanitization: The skill uses standard JSON encoding for payloads and defaults to a strict allowed_mentions policy ({"parse": []}) to prevent accidental or malicious mass pings (e.g., @everyone).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 03:36 PM
Security Audit — agent-trust-hub — discord