wolt

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow for this skill is to execute the wolt-pp-cli binary with user-supplied flags (e.g., menu show, pages search, venues-now), which fetches restaurant/menu content from Wolt’s public web/API and returns it as readable JSON that the agent then ingests into LLM context—i.e., outsider-authored free text/data from a public web source.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's prerequisite install commands fetch and execute remote code at runtime (npm package via "npx -y @mvanhorn/printing-press-library install wolt --cli-only" and the Go install path "github.com/mvanhorn/printing-press-library/.../wolt-pp-cli@latest"), and the skill depends on that fetched CLI binary to run, so these are runtime external dependencies that execute remote code.