build-feature

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

• Third-party content exposure detected (high risk: 1.00). The skill instructs the agent to call searchRecipes and getRecipe against the ShipSwift recipe server (https://api.shipswift.app/mcp) to fetch recipe source, architecture explanations, and integration checklists that the agent must read and use to drive coding and integration decisions, exposing it to untrusted third-party content that could inject instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires fetching ShipSwift recipes at runtime from the recipe MCP URL https://api.shipswift.app/mcp and injects those fetched recipe contents into the agent workflow to drive prompts and generated code, so this external endpoint directly controls agent instructions.