youtube-pipeline

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

• Third-party content exposure detected (high risk: 1.00). The skill's Phase 1 workflow explicitly instructs the user to export/copy YouTube Studio "Inspiration" outerHTML and run the included parse_yt_studio.ps1 (and to use the Google Trends CLI and competitor video links) so the agent will ingest and act on open/public third‑party content (YouTube/Google/competitor pages) as part of its decision-making pipeline.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.40). The skill instructs the agent to run local PowerShell and Node scripts (including an explicit PowerShell "-ExecutionPolicy Bypass" invocation, which bypasses a security control), but it does not ask for sudo/elevation, creation of users, or modification of system/service/SSH files, so the risk is present but limited to running user-level scripts and bypassing PowerShell policy.