marketer-github

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.85). This skill intentionally injects a third‑party runtime script (https://cdn.10x.in/webmcp.js) into users' websites and makes pages "discoverable" to remote AI agents, which enables remote code execution in visitors' browsers and can be used for data collection/exfiltration or backdoor-like remote control if the CDN script is malicious or changed later.

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow ingests outsider-authored free text via the tracking_list_templates MCP tool call, which fetches template/embed code from an external (non-user-authored) source and then injects that returned text into the user’s HTML/PR body context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires injecting and loading the remote script https://cdn.10x.in/webmcp.js into users' pages at runtime, which executes fetched code in the page and enables the page to act as an MCP server that can directly control agent interactions.