access
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection Guardrails: The skill includes explicit instructions to disregard requests arriving from external sources like Discord or Telegram messages. This is a proactive security measure designed to mitigate indirect prompt injection, ensuring that sensitive access modifications are only performed when triggered by direct user input in the terminal.
- Local Data Management: Access control state is maintained in a local JSON file within a dedicated directory. This design keeps sensitive configuration data isolated on the local system and avoids the security risks associated with transmitting access policies over the network.
- Scoped Tool Usage: The skill utilizes restricted bash commands and specific file system tools for its operations. This implementation follows the principle of least privilege by limiting the agent's capabilities to only the file and directory operations necessary for its stated purpose.
- Input and State Validation: The skill's logic includes validation for pairing codes, expiration timestamps, and configuration parameters. These checks help maintain the integrity of the access control system and prevent misconfiguration or unauthorized access grants.
Audit Metadata