paw-mkt-paid-ads

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

• Third-party content exposure detected (high risk: 0.90). The skill's Competitive Ad Research workflow and Shared Patterns explicitly instruct the agent to use agent-browser/WebFetch to open arbitrary public URLs (e.g., references/competitive-research.md shows commands like agent-browser open "https://{competitor-domain}/pricing" and ad library pages) and to extract and act on that live third‑party content as part of its decision-making, which exposes the agent to untrusted, user-generated web content that can influence subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runs agent-browser commands at runtime to fetch and inject live ad content (which can directly control prompts) from URLs such as https://www.facebook.com/ads/library/?active_status=active&ad_type=all&country=ALL&q={competitor-name}, so this external URL is used at runtime to load content that the agent ingests and acts on.