Silmaril
Protect model requests, responses and tool activity with Silmaril Firewall. Connect through your application, coding agents or AI platform.
Choose your path
Protect an AI application Use a language SDK or framework integration to check model and tool activity in your application.Deploy on macOS Protect local coding agents with a direct install or a managed rollout through your MDM.Connect an AI platform Connect Claude Enterprise, LiteLLM, Vercel AI Gateway, Bedrock AgentCore, Azure Foundry or Google Cloud.
Classify a request
Start with the quick start. Apply the returned decision with your chosen enforcement mode.
Language SDKs
Add checks where your application receives input, calls tools and returns model output.
AI platforms and gateways
Choose your platform for connection settings, model and tool routes, and enforcement.
Claude Enterprise
Connect conversations and tool calls through native inference hooks.
LiteLLM
Add pre-call and post-call guardrails to your proxy.
Vercel AI Gateway
Wrap gateway models with Silmaril middleware.
Bedrock AgentCore
Check model and MCP traffic through Gateway interceptors.
Azure Foundry
Configure Foundry model and tool routes in Azure API Management.
Google Cloud
Configure Agent Runtime model and tool routes through Agent Gateway.
Integrate and operate
Coding agentsSet up Codex, Claude Code, Pi, and other supported agent plugins.Framework SDKsAdd protection to Vercel AI SDK, LangChain, or LangGraph.Verify protectionConfirm traffic, connected boundaries, and blocking behavior.Firewall MCPLet your MCP client investigate findings and traffic.Self-hostingPlan a deployment in your cloud account with Silmaril.









