pp-apify

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to run and read results from Apify Actors that scrape public sites (e.g., "apify-pp run trudax/reddit-scraper", "apify-pp search ... --actors twitter,reddit,hn", and apify-pp store), which ingest untrusted, user-generated web content that the agent is expected to parse and act on (digest/publish/workflow), so third-party content can influence subsequent tool use and decisions.