# Silmaril > Security for agents that self-improves. Silmaril Firewall protects AI applications and agents. Silmaril provisions credentials and deployment endpoints during setup. - [Home](https://silmaril.security) - [Documentation](https://silmaril.security/docs) - [Documentation index](https://silmaril.security/docs/llms.txt) - [Full documentation text](https://silmaril.security/docs/llms-full.txt) - [Book a call](https://cal.com/silmaril/30min) ## Firewall / Start - [Overview](https://silmaril.security/docs): Connect your applications, coding agents and AI platforms to Silmaril Firewall. - [Get access](https://silmaril.security/docs/start/access): Get the credentials and endpoints for your Silmaril deployment. - [Quick start](https://silmaril.security/docs/start/quickstart): Install an SDK, classify a request, and verify your connection. - [Verify your integration](https://silmaril.security/docs/start/verify): Confirm traffic, connected boundaries, and enforcement before production rollout. ## Firewall / Integrations - [Framework integrations](https://silmaril.security/docs/integrations/frameworks): Choose an integration for Vercel AI SDK, LangChain, LangGraph, or Deep Agents. - [Vercel AI SDK](https://silmaril.security/docs/sdk/vercel-ai): Check model input and generated text before returning a Vercel AI SDK response. - [LangChain and LangGraph](https://silmaril.security/docs/sdk/langchain): Protect LangChain model calls and LangGraph execution in TypeScript or Python. - [Deep Agents](https://silmaril.security/docs/integrations/deep-agents): Protect root agents, tools, and subagents in TypeScript or Python. - [AI gateways](https://silmaril.security/docs/gateways/overview): Connect your AI platform or gateway to Silmaril. - [Claude Enterprise](https://silmaril.security/docs/integrations/claude-enterprise): Connect governed Claude conversations and tool calls through native Enterprise inference hooks. - [LiteLLM guardrails](https://silmaril.security/docs/integrations/litellm): Add Silmaril as a LiteLLM pre-call and post-call guardrail and validate coverage in warn mode. - [Amazon Bedrock AgentCore](https://silmaril.security/docs/integrations/bedrock-agentcore): Connect model and MCP traffic through native AgentCore Gateway interceptors. - [Microsoft Foundry and Azure AI gateway](https://silmaril.security/docs/integrations/azure-foundry): Configure model and tool boundaries for Microsoft Foundry workloads in Azure API Management. - [Google Cloud Agent Gateway](https://silmaril.security/docs/integrations/google-agent-gateway): Configure model and tool boundaries for Google Cloud Agent Runtime and Agent Gateway. - [Vercel AI Gateway](https://silmaril.security/docs/gateways/vercel): Check input and generated text for model calls routed through Vercel AI Gateway. - [Agent plugins](https://silmaril.security/docs/integrations/agent-plugins): Secure Codex, Claude Code, OpenClaw, Hermes, OpenCode, Pi, Cursor, GitHub Copilot CLI, and VS Code at the coding-agent boundary. ## Firewall / Guides - [Events and conversations](https://silmaril.security/docs/guides/events): Label execution boundaries and preserve ordered context across an agent sequence. - [Modes and enforcement](https://silmaril.security/docs/guides/enforcement): Choose how a verdict affects execution and how your application handles failed checks. - [Verdicts and outcomes](https://silmaril.security/docs/sdk/outcomes): Use prediction as the benign or malicious verdict, then route optional malicious detail with typed outcome constants. - [Review depth](https://silmaril.security/docs/guides/review-depth): Choose standard or deep review for a classification request. ## Firewall / Operate - [Review findings](https://silmaril.security/docs/operate/findings): Connect a Firewall finding to its event, decision, and application behavior. - [Firewall MCP](https://silmaril.security/docs/integrations/firewall-mcp): Connect an MCP client to read-only, tenant-scoped Silmaril Firewall evidence. - [Troubleshooting](https://silmaril.security/docs/operate/troubleshooting): Resolve access failures, missing activity, and classification errors. ## Firewall / Deployment - [Self-hosted deployments](https://silmaril.security/docs/deployment/self-hosting): Plan a Silmaril deployment in your cloud account with clear operational ownership. - [Data handling](https://silmaril.security/docs/deployment/data-handling): Understand the event data sent to Firewall and plan evidence access for your deployment. ## Firewall / SDK reference - [TypeScript SDK](https://silmaril.security/docs/sdk/typescript): Install the TypeScript SDK, classify user input and tool output, and route the verdict. - [Python SDK](https://silmaril.security/docs/sdk/python): Install the Python SDK, classify user input and tool output, and handle blocks with synchronous or asynchronous clients. - [Go SDK](https://silmaril.security/docs/sdk/go): Install the Go SDK, classify user input and tool output, and handle typed block errors. - [Java SDK](https://silmaril.security/docs/sdk/java): Java 0.7 guide for classifying user input and tool output and handling block verdicts. ## Firewall / Architecture - [Context and intent](https://silmaril.security/docs/concepts/intent-and-outcomes): Understand an action through the task, the user, the data, and what happened before it. - [Independent enforcement](https://silmaril.security/docs/concepts/independent-enforcement): Separate the agent's proposal, the security decision, and the action that follows. - [Learning your application](https://silmaril.security/docs/concepts/learning-your-application): A self-improving defense that learns your workflows and adapts as they change. ## macOS / Start - [Silmaril for macOS](https://silmaril.security/docs/macos): Protect local coding agents on one Mac or deploy through Jamf Pro, Iru, or Fleet. - [Install](https://silmaril.security/docs/macos/install): Set up Silmaril on one Mac using the Stable DMG. - [Verify protection](https://silmaril.security/docs/macos/verify): Check the release, connection, and current agent activity. ## macOS / Managed deployment - [Managed rollout](https://silmaril.security/docs/macos/managed-deployment): Prepare a configuration profile and package for deployment through MDM. - [Jamf Pro](https://silmaril.security/docs/macos/jamf): Deploy Silmaril with a device profile and a scoped package policy. - [Iru](https://silmaril.security/docs/macos/iru): Deploy Silmaril with a System Channel profile and an audit-and-enforce Custom App. - [Fleet](https://silmaril.security/docs/macos/fleet): Deploy Silmaril with a configuration profile and a custom software package. ## macOS / Configure - [Managed settings](https://silmaril.security/docs/macos/settings): Configure optional policy and look up the 19 supported MDM keys. - [Updates and channels](https://silmaril.security/docs/macos/updates): Choose app-owned or MDM-owned updates and verify each deployed build. ## macOS / Maintain - [Troubleshooting](https://silmaril.security/docs/macos/troubleshooting): Resolve connection, agent activity, and managed policy issues on a Mac. - [Uninstall](https://silmaril.security/docs/macos/uninstall): Remove Silmaril through MDM and resolve items reported in the uninstall receipt. ## API Reference / Start - [API Reference](https://silmaril.security/docs/api-reference/overview): Classify user input, tool calls, tool responses, and model output over HTTP. - [Authentication](https://silmaril.security/docs/api-reference/authentication): Authenticate Firewall API requests with your deployment API key. ## API Reference / Firewall - [Classify an event](https://silmaril.security/docs/api-reference/classify): Return a Firewall verdict for one complete event. - [Classify a batch](https://silmaril.security/docs/api-reference/batch): Return a Firewall verdict for each independent event in input order. ## Other pages - [Privacy](https://silmaril.security/privacy) - [Terms](https://silmaril.security/terms) - [Status](https://silmaril.security/status)