account-intel

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

• Third-party content exposure detected (medium risk: 0.30). In the required workflow, the agent actively fetches outsider-authored free text via external search/scraping APIs (YouTube search snippets from mpp.orthogonal.com, Brave news results, and Exa web citation contents) using user-provided --name/--handle/--query parameters, then ingests that text for synthesis.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly integrates payment rails and settlement calls: it uses Circle Gateway / SELAT Router nanopayments (x402, MPP on Tempo), compiles steps into selat-pay calls, and supports real "paid" runs that settle from a funded Circle Gateway balance. Those are specific payment/settlement APIs and on-chain settlements (including probe vs. --pay real settlement), so the skill can execute real financial transfers to upstream providers — i.e., direct financial execution authority.