Prismic Automation
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses a verified MCP endpoint (
https://rube.app/mcp) to provide Prismic CMS functionality. Analysis of the instructions and metadata revealed no evidence of obfuscation, credential harvesting, or unauthorized network activity. - [NO_CODE]: No executable files, scripts, or binary assets are packaged with the skill, mitigating risks associated with direct code execution or malicious payload delivery.
- [INDIRECT_PROMPT_INJECTION]: The skill's functionality involves retrieving untrusted data from the Prismic CMS (e.g., via
PRISMIC_CONTENT_API_QUERY_DOCUMENTSinSKILL.md). Although there are no specific boundary markers or sanitization steps described, the toolset is restricted to content retrieval and does not provide an agent with high-privilege capabilities, effectively limiting the risk of indirect injection.
Audit Metadata