# Edgee documentation

> Edgee is an Agent Gateway that cuts LLM costs through token compression and intelligent routing, with full visibility and control over agent usage.

- [Welcome to Edgee](https://www.edgee.ai/docs/introduction.md): Route agent requests, control spending, and understand team usage.
- [FAQ](https://www.edgee.ai/docs/introduction/faq.md): Answers about setup, billing, routing, data, and troubleshooting.
- [Start with your coding agent](https://www.edgee.ai/docs/quickstart/index.md): Install Edgee and its aliases, launch your coding agent as usual, and verify your first session.
- [Set up your team](https://www.edgee.ai/docs/quickstart/team.md): Invite a pilot group, connect their agents, assign a squad, and verify spending controls.
- [Claude Code](https://www.edgee.ai/docs/agents/claude-code.md): Use Edgee with Claude Code to cut token costs and extend sessions, with zero manual configuration.
- [Codex](https://www.edgee.ai/docs/agents/codex.md): Use Edgee with Codex to cut token costs and extend sessions, with zero manual configuration.
- [OpenCode](https://www.edgee.ai/docs/agents/opencode.md): Use Edgee with OpenCode for building AI applications directly in your terminal.
- [Cursor](https://www.edgee.ai/docs/agents/cursor.md): Use Edgee with Cursor to access any model through a single API key, with cost tracking and token compression.
- [GitHub Copilot CLI](https://www.edgee.ai/docs/agents/copilot-cli.md): Route GitHub Copilot CLI requests through Edgee using your Copilot subscription.
- [GitHub Copilot (VS Code)](https://www.edgee.ai/docs/agents/copilot-vscode.md): Route GitHub Copilot in VS Code through Edgee to cut token costs and gain full observability, with one command.
- [Xcode](https://www.edgee.ai/docs/agents/xcode.md): Route Xcode's Coding Assistant through Edgee to cut token costs and gain full observability over every session.
- [CodeBuddy](https://www.edgee.ai/docs/agents/codebuddy.md): Route CodeBuddy through Edgee to cut token costs and gain full observability, with one command.
- [Crush](https://www.edgee.ai/docs/agents/crush.md): Route Charm's Crush through Edgee to cut token costs and gain full observability, with one command.
- [Pi](https://www.edgee.ai/docs/agents/pi.md): Route the Pi coding agent through Edgee to cut token costs and gain full observability, with one command.
- [Kimi Code](https://www.edgee.ai/docs/agents/kimi.md): Route Moonshot AI's Kimi Code through Edgee to cut token costs and gain full observability, with one command.
- [Kilo Code](https://www.edgee.ai/docs/agents/kilo.md): Use Edgee with the Kilo Code CLI to route every terminal coding session through the gateway.
- [omp](https://www.edgee.ai/docs/agents/omp.md): Use Oh My Pi through Edgee with your usual terminal workflow.
- [GitHub Copilot in IntelliJ IDE (Beta)](https://www.edgee.ai/docs/agents/copilot-intellij.md): Use GitHub Copilot in IntelliJ IDE through Edgee.
- [GitHub Copilot Desktop (Beta)](https://www.edgee.ai/docs/agents/copilot-desktop.md): Route local GitHub Copilot desktop sessions through Edgee on macOS.
- [Claude Desktop (Beta)](https://www.edgee.ai/docs/agents/claude-desktop.md): Use Claude Code inside Claude Desktop through Edgee.
- [Codex Desktop (Beta)](https://www.edgee.ai/docs/agents/codex-desktop.md): Use the Codex tab in the ChatGPT desktop app through Edgee.
- [Edgee CLI](https://www.edgee.ai/docs/features/cli.md): Launch agents, manage your login and organization, configure settings, and inspect usage.
- [Authentication, organizations, and profiles](https://www.edgee.ai/docs/cli/authentication.md): Keep work and personal accounts separate and select where your agent usage is recorded.
- [Aliases and desktop shortcuts](https://www.edgee.ai/docs/cli/aliases.md): Start agents through Edgee using their usual commands, and undo that setup when needed.
- [Claude Code statusline](https://www.edgee.ai/docs/cli/statusline.md): Install, check, and repair the Edgee usage indicator in Claude Code.
- [CLI configuration](https://www.edgee.ai/docs/cli/configuration.md): Understand local credentials, endpoint overrides, and debugging preferences.
- [Apps and additional agents](https://www.edgee.ai/docs/cli/apps.md): Find launch targets supported by your CLI version and verify an editor is routed through Edgee.
- [Token Compression](https://www.edgee.ai/docs/features/token-compression.md): Token compression is the surgical removal of redundancy, not summarization. Edgee treats it in two distinct layers, input and output.
- [Distribute plugins to your team](https://www.edgee.ai/docs/features/plugins.md): Create shared skills and tools, assign them to members, and verify delivery through the CLI.
- [Organization](https://www.edgee.ai/docs/features/organization.md): Select your organization, configure shared settings, and understand how they apply to squads and agents.
- [Single sign-on (SSO)](https://www.edgee.ai/docs/features/sso.md): Connect an OIDC identity provider and verify sign-in for an organization member.
- [Manage your team](https://www.edgee.ai/docs/features/team.md): Invite members, assign seats, enroll agents, and inspect per-member usage.
- [Manage squads](https://www.edgee.ai/docs/features/squads.md): Group members, apply shared settings, and track a team budget.
- [Observability](https://www.edgee.ai/docs/features/observability.md): Track every request, measure every token, optimize every dollar.
- [Sessions](https://www.edgee.ai/docs/features/observability/sessions.md): Every coding session, captured. Inspect tokens, costs, and compression — and share the report with one link.
- [Debug Logs](https://www.edgee.ai/docs/features/debug.md): Enable debug mode for your coding agent and inspect its requests, responses, and compression results.
- [Data destinations](https://www.edgee.ai/docs/features/integrations/data-destinations.md): Forward hosted-gateway usage logs to your own observability endpoint, in addition to Edgee's internal pipeline.
- [Connect GitHub and track repository activity](https://www.edgee.ai/docs/features/integrations/github.md): Link repositories to your organization and verify session attribution to repositories and pull requests.
- [Bring Your Own Keys (BYOK)](https://www.edgee.ai/docs/features/byok.md): Use your own provider keys for billing and provider-side controls, while keeping Edgee's routing and observability.
- [Model & provider access](https://www.edgee.ai/docs/features/model-access.md): Control which models and which providers your organization is allowed to route to.
- [Retry, Fallback & Reroute](https://www.edgee.ai/docs/features/retry-and-fallback.md): Automatic retry, provider fallback, and unconditional model reroute for resilient LLM routing.
- [Routing strategies](https://www.edgee.ai/docs/features/routing-strategies.md): Choose models and fallbacks, then change routing as a budget is consumed.
- [Usage limits & budgets](https://www.edgee.ai/docs/features/usage-limits.md): Cap coding-agent spend per member or squad, and application spend per standard API key.
- [Alerts and budget monitors](https://www.edgee.ai/docs/features/alerts.md): Create spending monitors and choose how your team receives notifications.
- [Self-Hosting Overview](https://www.edgee.ai/docs/self-hosting/overview.md): Run the Edgee AI Gateway on your own infrastructure — connected or air-gapped.
- [Docker Compose](https://www.edgee.ai/docs/self-hosting/docker-compose.md): Run the Edgee AI Gateway with Docker Compose, connected or headless.
- [Kubernetes & Helm](https://www.edgee.ai/docs/self-hosting/kubernetes-helm.md): Install the Edgee AI Gateway on Kubernetes with the public Helm chart.
- [Headless configuration](https://www.edgee.ai/docs/self-hosting/configuration.md): Configure a licensed gateway from local files and verify a model request.
- [Troubleshooting](https://www.edgee.ai/docs/troubleshooting.md): Resolve missing sessions, blocked requests, locked settings, and CLI launch issues.
- [LLM Router and SDKs](https://www.edgee.ai/docs/llm-router/index.md): Connect an application to Edgee using an existing client, an Edgee SDK, or HTTP.
- [Connect your application](https://www.edgee.ai/docs/llm-router/quickstart.md): Create a Gateway key, connect your client, and verify a model request.
- [Create an API Key](https://www.edgee.ai/docs/llm-router/quickstart/api-key.md): Create a new API key to use Edgee services.
- [Integrate Edgee to your stack](https://www.edgee.ai/docs/llm-router/quickstart/integration.md): Learn how to integrate Edgee to your stack and start using Edgee services in your application.
- [OpenAI SDK](https://www.edgee.ai/docs/llm-router/integrations/openai.md): Using Edgee with OpenAI SDK
- [Anthropic SDK](https://www.edgee.ai/docs/llm-router/integrations/anthropic.md): Use Edgee with the Anthropic SDK for building AI applications with Claude models.
- [LangChain](https://www.edgee.ai/docs/llm-router/integrations/langchain.md): Use Edgee with LangChain for building AI applications with chains, agents, and RAG.
- [OpenClaw](https://www.edgee.ai/docs/llm-router/integrations/openclaw.md): Use Edgee with OpenClaw to compress tokens, route requests, and observe every session.
- [TypeScript SDK](https://www.edgee.ai/docs/llm-router/sdk/typescript/index.md): Integrate the TypeScript SDK in your application.
- [TypeScript SDK Configuration](https://www.edgee.ai/docs/llm-router/sdk/typescript/configuration.md): Learn how to configure and instantiate the Edgee TypeScript SDK.
- [TypeScript SDK - Send Method](https://www.edgee.ai/docs/llm-router/sdk/typescript/send.md): Complete guide to the send() method in the TypeScript SDK.
- [TypeScript SDK - Stream Method](https://www.edgee.ai/docs/llm-router/sdk/typescript/stream.md): Complete guide to the stream() method in the TypeScript SDK.
- [TypeScript SDK - Tools (Function Calling)](https://www.edgee.ai/docs/llm-router/sdk/typescript/tools.md): Complete guide to function calling with the TypeScript SDK.
- [Python SDK](https://www.edgee.ai/docs/llm-router/sdk/python/index.md): Integrate the Python SDK in your application.
- [Python SDK Configuration](https://www.edgee.ai/docs/llm-router/sdk/python/configuration.md): Learn how to configure and instantiate the Edgee Python SDK.
- [Python SDK - Send Method](https://www.edgee.ai/docs/llm-router/sdk/python/send.md): Complete guide to the send() method in the Python SDK.
- [Python SDK - Stream Method](https://www.edgee.ai/docs/llm-router/sdk/python/stream.md): Complete guide to the stream() method in the Python SDK.
- [Python SDK - Tools (Function Calling)](https://www.edgee.ai/docs/llm-router/sdk/python/tools.md): Complete guide to function calling with the Python SDK.
- [Rust SDK](https://www.edgee.ai/docs/llm-router/sdk/rust/index.md): Integrate the Rust SDK in your application.
- [Rust SDK Configuration](https://www.edgee.ai/docs/llm-router/sdk/rust/configuration.md): Learn how to configure and instantiate the Edgee Rust SDK.
- [Rust SDK - Send Method](https://www.edgee.ai/docs/llm-router/sdk/rust/send.md): Complete guide to the send() method in the Rust SDK.
- [Rust SDK - Stream Method](https://www.edgee.ai/docs/llm-router/sdk/rust/stream.md): Complete guide to the stream() method in the Rust SDK.
- [Rust SDK - Tools (Function Calling)](https://www.edgee.ai/docs/llm-router/sdk/rust/tools.md): Complete guide to function calling with the Rust SDK.
- [Go SDK](https://www.edgee.ai/docs/llm-router/sdk/go/index.md): Integrate the Go SDK in your application.
- [Go SDK Configuration](https://www.edgee.ai/docs/llm-router/sdk/go/configuration.md): Learn how to configure and instantiate the Edgee Go SDK.
- [Go SDK - Send Method](https://www.edgee.ai/docs/llm-router/sdk/go/send.md): Complete guide to the Send() method in the Go SDK.
- [Go SDK - Stream Method](https://www.edgee.ai/docs/llm-router/sdk/go/stream.md): Complete guide to the stream() method in the Go SDK.
- [Go SDK - Tools (Function Calling)](https://www.edgee.ai/docs/llm-router/sdk/go/tools.md): Complete guide to function calling with the Go SDK.
- [Manage Gateway API keys](https://www.edgee.ai/docs/llm-router/api-keys.md): Create, restrict, inspect, and revoke the credentials used for model requests.
- [Application observability](https://www.edgee.ai/docs/llm-router/observability.md): Read response usage, inspect compression metrics, and tag application requests.
- [API overview](https://www.edgee.ai/docs/llm-router/api-reference/index.md): Choose the Gateway API for model requests or the Console API for organization management.
- [Authentication](https://www.edgee.ai/docs/llm-router/api-reference/authentication.md): How to authenticate to the Edgee API
- [Errors](https://www.edgee.ai/docs/llm-router/api-reference/errors.md): How Edgee API responds when errors occur.
- [OpenAI Chat Completion](https://www.edgee.ai/docs/llm-router/api-reference/chat-completion.md): Create chat completions using the OpenAI Chat Completion API format
- [OpenAI Responses](https://www.edgee.ai/docs/llm-router/api-reference/responses.md): Create responses using the OpenAI Responses API format
- [Anthropic Messages](https://www.edgee.ai/docs/llm-router/api-reference/messages.md): Create messages using Anthropic Messages API format
- [Models](https://www.edgee.ai/docs/llm-router/api-reference/models.md): List all available models in the Edgee AI Gateway
- [Count Tokens](https://www.edgee.ai/docs/llm-router/api-reference/count-tokens.md): Estimate token count for a set of messages without making an LLM call
- [Compress](https://www.edgee.ai/docs/llm-router/api-reference/compress.md): Compress a request payload without calling any LLM provider
- [Setup & Authentication](https://www.edgee.ai/docs/api-reference/console-api-setup.md): Getting started with the Edgee Console API - create tokens and authenticate requests
- [Export Cost Data](https://www.edgee.ai/docs/api-reference/console-api-export-logs.md): Export AI Gateway usage and cost data
- [Manage Gateway API Keys](https://www.edgee.ai/docs/api-reference/console-api-manage-gateway-api-keys.md): Create, list, update, and delete AI Gateway API keys
- [BYOK Setup](https://www.edgee.ai/docs/api-reference/console-api-provider-keys.md): Create and manage custom provider keys (BYOK)

## OpenAPI Specs

- [openapi](/docs/api-reference/openapi.json)
- [package-lock](/docs/package-lock.json)
- [package](/docs/package.json)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.