VTEX Agentic CX Analytics is an MCP (Model Context Protocol) server developed by Weni AI. It provides real-time analytics tools for VTEX Agentic CX chatbot projects, including conversation metrics, customer satisfaction scores (CSAT), Net Promoter Score (NPS), topic distribution, sales funnel data, and human support monitoring.
This policy describes what data the server accesses, how it is handled, and what information may be returned to the AI model or user interface.
When a tool is called, the server queries the Weni Insights Engine API on behalf of the authenticated user. The data accessed includes:
Some tools return data that may include personal identifiers. The following anonymization rules are applied before any data is returned to the model or user:
Contact #a3f2b1) using a one-way SHA-256 hash.
The original name is never exposed.a***@company.com). Only the first character and domain are visible.All requests to the server are authenticated via OpenID Connect (OIDC) using Keycloak. The server does not store authentication tokens or credentials. Access tokens are used only to authorize API requests during the active session and are never logged or persisted.
The MCP server does not store, log, or retain any user data, conversation content, or analytics results. All data is fetched in real time from the Weni Insights Engine API and returned directly to the requesting client. No caching of personal data occurs.
The server communicates with the following external service:
insights-engine.weni.ai) —
the source of all analytics data. Data handling by this service is governed by
Weni's Privacy Policy.Users whose data appears in the analytics (e.g., support agents, end customers) have rights under applicable data protection laws, including the Brazilian General Data Protection Law (LGPD) and GDPR where applicable. To exercise these rights, contact privacy@weni.ai.
For questions about this privacy policy or data practices, contact:
Weni AI
privacy@weni.ai
https://weni.ai