env-workflow-vercel

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

• Third-party content exposure detected (high risk: 0.80). The skill instructs the agent to fetch a recipe from a public third-party site (recipe://fullstackrecipes.com and via curl to https://fullstackrecipes.com/api/recipes/env-workflow-vercel), which the agent would retrieve and interpret as part of its workflow, exposing it to untrusted web content that could carry injected instructions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill includes a runtime fetch (curl) to retrieve recipe content from https://fullstackrecipes.com/api/recipes/env-workflow-vercel which would supply operational instructions used by the agent, so this external URL can directly control prompts/behavior.