pp-luma

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md states the CLI “reads Luma’s public, read-only discovery API” and mirrors fetched event/calendar details into a local SQLite store, which implies runtime ingestion of outsider-authored free text from a public web API into the agent’s LLM context via the CLI’s JSON/stdout output in --agent mode.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's prerequisites include direct install commands that fetch and execute remote code (go install github.com/mvanhorn/printing-press-library/library/social-and-messaging/luma/cmd/luma-pp-cli@latest and go install github.com/mvanhorn/printing-press-library/library/social-and-messaging/luma/cmd/luma-pp-mcp@latest), which are required to run the skill and therefore pose a runtime external-code execution dependency.