tailor-resume

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

• Third-party content exposure detected (high risk: 0.95). Step 1 uses Claude in Chrome MCP tools to fetch a user-provided job URL and then parses/extracts the job posting text (outsider-authored free text from the employer’s page) into posting.md and into the agent’s LLM context for tailoring.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly fetches and parses a user-supplied job posting URL at runtime (the job URL passed in $ARGUMENTS, e.g., any external https://... job posting) using the Chrome MCP navigate/get_page_text flow and then uses that page content to drive resume-generation logic, so arbitrary external pages could directly influence prompts and outputs (prompt-injection risk).