workflow

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFE
Full Analysis
  • Vendor-Maintained Package Installation: The skill recommends installing workflow and @workflow/ai from NPM and running npx workflow@latest. These are official resources associated with the skill's authoring organization. Usage of these packages is standard for building with the Vercel Workflow ecosystem.
  • Credential Handling via Official CLI: The documentation instructs users to run vercel env pull to retrieve environment variables like VERCEL_OIDC_TOKEN. This is the recommended secure method for managing credentials within the Vercel platform environment.
  • Agent Input Surface: The DurableAgent implementation processes external inputs (such as search queries and user messages) to drive tool execution and LLM responses. While this creates a surface for indirect prompt injection, it is the primary intended functionality of an AI agent skill. Developers should implement standard sanitization and validation for data passed into agent tools.
  • Proactive Security Disclosure: The skill includes an advisory regarding a patched vulnerability (CVE GHSA-9r75-g2cr-3h76) and correctly instructs users to upgrade to safe versions (>=4.2.0-beta.64), demonstrating a proactive security posture.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 06:29 PM