aisa

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

• Third-party content exposure detected (low risk: 0.10). The AIsa skill searches and fetches data from external web, company, or social tools via the AIsa tool registry, which can ingest outsider-authored content upon user search requests.

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

• Direct money access detected (high risk: 0.70). The skill describes invoking paid tools via an AIsa platform CLI and MCP (quote and call), which involves executing paid tool calls and spending funds within an authorized scope, constituting a direct financial operation (purchases/charges). While credentials and auth sessions are mentioned, no independent, service-enforced access control validating the specific financial transaction (such as a distinct checkout or server-side spending approval gate independent of the agent's stored keys) is established for the call route.