mastra-framework-guide

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs using remote documentation via WebFetch (e.g., fetching https://mastra.ai/llms.txt and other .md pages) which is outsider-authored free text that can be ingested into the agent’s LLM context at runtime.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent to fetch remote documentation at runtime (e.g., https://mastra.ai/llms.txt and specific .md pages like https://mastra.ai/reference/workflows/workflow-methods/then.md) and even shows a WebFetch({ url, prompt }) usage, meaning fetched content would directly guide agent prompts and behavior.