suede-ai-seo
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves analyzing external web content and robots.txt files, which exposes it to potential indirect prompt injection from malicious websites. However, the instructions include specific safeguards, such as requiring the use of host-approved read-only tools, prohibiting the use of credentials or session cookies in requests, and enforcing a structured 'Output Contract' to prevent the agent from treating data as instructions.
- [SAFE]: The skill adheres to security best practices by defining clear boundaries for network operations and workspace access. It references reputable technology organizations and documented industry research for its technical guidance, and it appropriately defers to other specialized tools for general SEO tasks, ensuring a narrow and safe operational scope.
Audit Metadata