pp-scrape-do

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is to run scrape-do-pp-cli (e.g., google, scrape, batch) which can fetch and return scraped web content/SERPs at runtime; that returned free-form text/JSON is then ingested into the agent’s LLM context via the CLI output (outsider-authored public web content path).