vercel-cli-with-tokens

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Deploying from a Remote Repository" and related steps explicitly instruct cloning a repository with "git clone " and checking repository files (package.json, vercel.json), which means the agent would fetch and ingest arbitrary user-generated code/metadata from external repos that can alter deployment/build behavior.