glc PromptGuard
glc PromptGuard provides prompt-injection defense for AI agents and RAG pipelines, checking text for intent, source context, and impact.
Open productWhat it does
About glc PromptGuard
glc PromptGuard is a multilingual prompt-injection defense service available on the GLC MCP platform. Designed for AI agents and RAG pipelines, it evaluates untrusted text inputs before main LLM calls to prevent unauthorized instruction overrides. The service inspects text by analyzing intent, source context, and impact, returning an injection flag, score, intent, and policy outcome. Developers can adjust evaluation policies based on context options like user prompts, RAG chunks, or tool results, and interact with the service through REST or streamable HTTP MCP endpoints.
- Availability:
- launched
- Access:
- freemium
- Format:
- service
- Platform:
- api
Pricing at launch: Includes 100 free credits per account. Checks consume 1 credit base rate plus LLM usage.
Core capabilities
Untrusted Text Injection Inspection
Evaluates untrusted text for prompt injection risks, returning an injection flag, evaluation score, detected intent, and policy output.
Context-Aware Policy Adjustment
Keep discovering
Related debuts
Continue with products that solve a similar problem or serve the same kind of maker.
