mastra
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as an educational and operational guide for the Mastra framework. It contains detailed instructions for using the
mastraCLI, REST APIs, and project templates. - [EXTERNAL_DOWNLOADS]: The skill references several external resources, all of which are official vendor assets:
- Documentation at
https://mastra.ai/llms.txtandhttps://mastra.ai/docs/*. - Platform APIs at
https://platform.mastra.ai/andhttps://observability.mastra.ai. - Official GitHub repository at
https://github.com/mastra-ai/skills. - These are verified vendor resources from
mastra-aiand are considered safe. - [COMMAND_EXECUTION]: The skill provides numerous CLI command examples for project setup (
npm create mastra@latest), deployment (mastra deploy), and API interaction (mastra api agent list). These are standard operations for a development framework and do not include malicious payloads. - [DATA_EXPOSURE]: The documentation correctly advises users on secret management, explicitly warning against committing secrets or printing them to logs. It recommends using
.envfiles and environment variables, which is a standard security best practice. - [REMOTE_CODE_EXECUTION]: While the skill mentions
PlatformSandboxfor remote command execution, this is presented as a native capability of the Mastra Platform for isolated task execution, with instructions on how to initialize and destroy sandboxes securely. - [DYNAMIC_CONTEXT_INJECTION]: No use of the
!commandsyntax was found in theSKILL.mdor other markdown files.
Audit Metadata