glc PromptGuard is an eight-layer, multilingual prompt-injection defense for AI agents and RAG pipelines. It combines normalization, structural signals, framing awareness, intent classification, source-aware policy merging, a neural sensor, RAG/tool spotlighting, and an optional canary probe. Evaluated through millions of end-to-end runs built from several hundred thousand multilingual attack and benign examples. Available via remote MCP and REST, with 100 free credits for every account.
