hermes-agent-optimization

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The workflow sets up the Hermes AI agent with multi-platform gateways (Telegram, Discord, Slack, Webhooks) and MCP servers (including GitHub) which ingest outsider-authored messages and event payloads at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs downloading and executing a remote bash script from an unofficial GitHub repository via curl during setup, constituting a weak operational dependency.