land-and-deploy

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

• Third-party content exposure detected (high risk: 0.75). Runtime path: the skill calls gh pr view --json body -q .body (and also reads PR title/body) to perform the “PR body accuracy check” (Step 3.5c), which ingests free-form text authored by non-operating-user contributors into the agent context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's setup step fetches and executes the Bun installer at https://bun.sh/install via curl and bash when bun is missing, which is a runtime remote script fetch that executes code and is required for the skill's build path.