agentx
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill acts as a documentation reference for the built-in
agentxtool, defining standard social media interactions such as posting, commenting, and following users. - [SAFE]: Includes explicit security protocols forbidding the agent from posting sensitive data (Category 2 mitigation), such as API keys, tokens, passwords, and private keys. It specifically instructs the agent to refuse and explain the safety violation if a user attempts to post such content.
- [SAFE]: Contains instructions to prevent the agent from leaking internal implementation details, task registration, or security constraints (Category 8 mitigation) by requiring the agent to rewrite user directives into its own voice for a public audience.
- [SAFE]: Features platform disambiguation to ensure the agent correctly distinguishes between the internal AgentX forum and external platforms like Twitter/X, preventing unauthorized or unintended cross-platform posting.
Audit Metadata