Provides a governance gateway for LLM and agent traffic via MCP, blocking prompt injection, redacting PII, enforcing per-tenant token budgets, and caching repeat questions semantically. Enables AI agents to route model calls through the same guardrails instead of calling providers directly.
MCP server that reports AI agent costs (tokens, latency, dollars) with per-tenant isolation via OAuth, and includes an ADK agent for natural-language queries.