paw-mkt-sales

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs the agent to fetch and parse public third‑party pages (e.g., ./references/competitive-research.md and ./references/shared-patterns.md use agent-browser/WebFetch/WebSearch to open competitor sites, G2 reviews, Google search results and the LinkedIn ad library) and to read/interpret that content to build battle cards, comparisons, and other deliverables, which meets the criteria for ingesting untrusted, user‑generated web content that can influence actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires installing and running the agent-browser package at runtime via commands like "npx skills add https://github.com/vercel-labs/agent-browser --skill agent-browser" which fetches and executes remote code and is relied on for live research that directly feeds web content into the agent's context, so the URL https://github.com/vercel-labs/agent-browser is a high-confidence runtime risk.