cloudflare

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

• Third-party content exposure detected (high risk: 0.70). This skill routinely calls the Cloudflare API (cf-api.sh -> https://api.cloudflare.com, and scripts like scripts/analytics.sh fetching /zones/{zone_id}/security/events and scripts/workers.sh fetching worker scripts) and parses/prints responses that include untrusted, user-generated data (request paths, user agents, analytics/events, and worker script contents) which the agent reads/interprets as part of its workflow, creating a risk of indirect prompt injection.