contentful-personalization

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a bash wrapper (scripts/run) to execute its core logic via a local Node.js binary (contentful-personalization.mjs). This is the standard operational model for the workflow engine it employs.
  • [EXTERNAL_DOWNLOADS]: The onboard sub-skill is designed to automate the installation of official Contentful and Ninetailed SDK packages (e.g., @ninetailed/experience.js, @contentful/optimization) via package managers. These are trusted vendor-provided libraries.
  • [DATA_EXFILTRATION]: The doctor and onboard sub-skills explore the project codebase and environment variables (e.g., CONTENTFUL_TOKEN, NEXT_PUBLIC_NINETAILED_CLIENT_ID) to verify configuration. This data access is used strictly for local diagnostics and setup verification.
  • [PROMPT_INJECTION]: The live-debug functionality ingests content from external URLs using Chrome DevTools. While this presents an indirect prompt injection surface where a malicious webpage could attempt to influence the agent, the risk is inherent to the debugging purpose and is mitigated by the tool's limited diagnostic scope.
  • [SAFE]: No signs of malicious obfuscation, credential theft, or unauthorized persistence mechanisms were found. The skill's behavior aligns with its documented purpose as a developer utility for Contentful services.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 02:54 AM
Security Audit — agent-trust-hub — contentful-personalization