upwork-application-session

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly opens and probes public Upwork job pages (see commands like node ../../scripts/upwork-search-inspect.mjs '<search-url-or-keyword>' and node ../../scripts/upwork-apply-probe.mjs '<job-url>'), reads client-written job posts/questions, and uses that content to decide actions and to draft/submit proposals, exposing the agent to untrusted third‑party input that could inject instructions.