azure-compute
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [External Data Ingestion]: The skill retrieves information from technical documentation repositories to provide up-to-date VM recommendations. This pattern is standard for providing accurate technical guidance but involves processing external content within the agent's context.
- [Public IP Detection]: The workflow utilizes a well-known service to determine the user's public IP address. This information is used to automatically configure Network Security Group (NSG) rules to limit access to the user's specific location, which is a recommended security practice for protecting management ports.
- [SSH Public Key Access]: The skill attempts to locate the user's public SSH key (
~/.ssh/id_rsa.pub) to facilitate VM provisioning. This uses the public part of the key pair, which is designed for sharing, and the skill follows best practices by using placeholders rather than literal key content in generated artifacts. - [Automated Command Execution]: The skill uses local tools like
git,gh, andazto automate tasks such as infrastructure provisioning and Pull Request creation. These actions are fundamental to the skill's purpose and include safeguards like using environment variables for passwords to avoid hardcoding secrets in configuration files.
Audit Metadata