pp-lever

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md, the runtime workflow uses the Lever Public postings API (https://api.lever.co/v0/postings/{company}) via the CLI to list/get company postings, meaning outsider-authored job-board content (e.g., company posting text) is ingested and surfaced to the agent.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill includes explicit install-time commands that fetch and execute remote code (e.g., "go install github.com/mvanhorn/printing-press-library/library/job-boards/lever/cmd/lever-pp-cli@latest" and the module reference "github.com/mvanhorn/printing-press-library/library/job-boards/lever/cmd/lever-pp-cli"), which are required dependencies and will run code fetched from external sources if the agent installs the CLI.