cover-letter

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow (SKILL.md Step 1) explicitly uses Claude-in-Chrome MCP tools to navigate to arbitrary job URLs and run get_page_text to extract full job postings from the public web, which the agent then reads and uses to compose tailored cover letters—exposing it to untrusted third-party content that can influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches arbitrary job posting URLs at runtime (e.g., https://example.com/jobs/vp-growth) via the Claude-in-Chrome MCP flow (navigate -> get_page_text) and injects that external page text into the agent context to drive the cover-letter generation, so the external URL content directly controls prompts.