pp-browserbase

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow described in SKILL.md’s Agent Mode/Automatic learning ingests outsider-authored free text at runtime from the user’s question (passed to browserbase-pp-cli recall "<user's question>" --agent and later used for discovery/teach/playbook logic), which can enable indirect prompt-injection-like behavior through that text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's prerequisites include go install commands that fetch and build remote code at runtime and are required for the CLI (e.g., "go install github.com/mvanhorn/printing-press-library/library/cloud/browserbase/cmd/browserbase-pp-cli@latest"), which executes remote content.