# Silmaril

> Security for agents that self-improves.

Silmaril Firewall protects AI applications and agents. Credentials and deployment endpoints are provisioned by Silmaril. Access is not self-serve.

- [Home](https://silmaril.dev)
- [Documentation](https://silmaril.dev/docs)
- [Documentation index](https://silmaril.dev/docs/llms.txt)
- [Full documentation text](https://silmaril.dev/docs/llms-full.txt)
- [Book a call](https://cal.com/silmaril/30min)

## Start
- [Overview](https://silmaril.dev/docs): Protect AI applications and agents at the points where they receive input, call tools, and produce output.
- [Quick start](https://silmaril.dev/docs/start/quickstart): Get access, install an SDK, classify your first request, and handle the verdict.

## Concepts
- [Intent and outcomes](https://silmaril.dev/docs/concepts/intent-and-outcomes): Understand an action through the task, the user, the data, and what happened before it.
- [Independent enforcement](https://silmaril.dev/docs/concepts/independent-enforcement): Separate the agent's proposal, the security decision, and the action that follows.
- [Learning your application](https://silmaril.dev/docs/concepts/learning-your-application): A self-improving defense that learns your workflows and adapts as they change.

## SDKs
- [TypeScript SDK](https://silmaril.dev/docs/sdk/typescript): Install the TypeScript SDK, classify user input and tool output, and route the verdict.
- [Python SDK](https://silmaril.dev/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.dev/docs/sdk/go): Install the Go SDK, classify user input and tool output, and handle typed block errors.
- [Java SDK](https://silmaril.dev/docs/sdk/java): Java 0.7 guide for classifying user input and tool output and handling block verdicts.
- [Framework SDKs](https://silmaril.dev/docs/integrations/frameworks): Choose a Silmaril integration for Vercel AI SDK, LangChain, or LangGraph.
- [Vercel AI SDK](https://silmaril.dev/docs/sdk/vercel-ai): Check model input and generated text before returning a Vercel AI SDK response.
- [LangChain and LangGraph](https://silmaril.dev/docs/sdk/langchain): Protect LangChain callbacks and Deep Agents graphs in TypeScript or Python.
- [Firewall outcomes](https://silmaril.dev/docs/sdk/outcomes): Use prediction as the benign or malicious verdict, then route optional malicious detail with typed outcome constants.

## AI gateways
- [AI gateways](https://silmaril.dev/docs/gateways/overview): Protect model traffic routed through LiteLLM or Vercel AI Gateway.
- [LiteLLM guardrails](https://silmaril.dev/docs/integrations/litellm): Add Silmaril as a LiteLLM pre-call and post-call guardrail and validate coverage in warn mode.
- [Vercel AI Gateway](https://silmaril.dev/docs/gateways/vercel): Check input and generated text for model calls routed through Vercel AI Gateway.

## Integrations
- [Agent plugins](https://silmaril.dev/docs/integrations/agent-plugins): Secure Codex, Claude Code, OpenClaw, Hermes, and OpenCode at the coding-agent boundary.
- [Firewall MCP](https://silmaril.dev/docs/integrations/firewall-mcp): Connect an MCP client to read-only, tenant-scoped Silmaril Firewall evidence.

## Deployment
- [Self-hosting](https://silmaril.dev/docs/deployment/self-hosting): Run Silmaril in a cloud account and region you control, next to the workload it protects.

## MacOS
- [Silmaril for macOS](https://silmaril.dev/docs/macos): Protect local coding agents on one Mac or deploy through Jamf Pro, Iru, or Fleet.
- [Install](https://silmaril.dev/docs/macos/install): Set up Silmaril on one Mac using the Stable DMG.
- [Managed rollout](https://silmaril.dev/docs/macos/managed-deployment): Prepare a configuration profile and package for deployment through MDM.
- [Jamf Pro](https://silmaril.dev/docs/macos/jamf): Deploy Silmaril with a device profile and a scoped package policy.
- [Iru](https://silmaril.dev/docs/macos/iru): Deploy Silmaril with a System Channel profile and an audit-and-enforce Custom App.
- [Fleet](https://silmaril.dev/docs/macos/fleet): Deploy Silmaril with a configuration profile and a custom software package.
- [Managed settings](https://silmaril.dev/docs/macos/settings): Configure optional policy and look up the 17 supported MDM keys.
- [Verify protection](https://silmaril.dev/docs/macos/verify): Check the app, connection, and live activity before expanding a rollout.
- [Uninstall](https://silmaril.dev/docs/macos/uninstall): Remove Silmaril through MDM and resolve items reported in the uninstall receipt.
- [Troubleshooting](https://silmaril.dev/docs/macos/troubleshooting): Resolve managed policy, connection, and protection issues on a Mac.

## Start
- [API Reference](https://silmaril.dev/docs/api-reference/overview): Classify user input, tool calls, tool responses, and model output over HTTP.
- [Authentication](https://silmaril.dev/docs/api-reference/authentication): Authenticate Firewall API requests with your deployment API key.

## Firewall
- [Classify an event](https://silmaril.dev/docs/api-reference/classify): Return a Firewall verdict for one complete event.
- [Classify a batch](https://silmaril.dev/docs/api-reference/batch): Return a Firewall verdict for each independent event in input order.

## Other pages
- [Privacy](https://silmaril.dev/privacy)
- [Terms](https://silmaril.dev/terms)
- [Status](https://silmaril.dev/status)