paw-mkt-social

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

• Third-party content exposure detected (high risk: 0.90). The skill's shared-patterns.md and agent-browser setup explicitly require "Context C — Live Website URL Audit" and show agent-browser/WebFetch commands to open public social pages (e.g., LinkedIn, X/Twitter, Facebook Ads Library), so the agent will fetch and interpret untrusted public web/social content as part of its required workflow and use that content to drive audits, strategy, and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs runtime use of agent-browser to open and fetch external pages (e.g., https://www.linkedin.com/search/results/people/?keywords=marketing), which will pull remote page text into the agent context and can directly control prompts/instructions.