Prompt-injection firewall for AI agents — scan untrusted text before LLM calls.
No remote MCP endpoint is published for this listing.
No capability manifest has been published for this listing yet.