What AI models does the agent use?

The agent has access to Claude, DeepSeek, Gemini, and other top models. It automatically routes to the best model for each task — you never have to think about API keys or token limits.

Yes. Each agent runs in an isolated environment. We don't train on your data, and you can export or delete everything at any time. The underlying engine (GoGogot) is open-source, so you can audit exactly how it works.

What can the agent actually do?

Browse the web, process files (CSV, PDF, images), run scheduled tasks, remember context across sessions, and communicate via Telegram or Slack. Think of it as a capable junior employee, not a chatbot.

How does memory work?

The agent maintains persistent memory across sessions. It remembers your preferences, past conversations, and task context. You can also explicitly tell it to remember or forget things.

Can I self-host the agent?

Absolutely. GoGogot is 100% open-source. Run it on your own hardware for free. cowork.ink is the managed version — we handle servers, model costs, updates, and uptime so you don't have to.

Is there a free trial?

We offer a 7-day money-back guarantee. Spin up an agent, give it real tasks, and if it doesn't save you time, we'll refund you — no questions asked.

What is the difference between AI middleware and an API gateway?

An API gateway handles authentication, rate limiting, and routing for HTTP APIs. AI middleware does all of that plus LLM-specific concerns: model routing, prompt sanitization, token cost tracking, semantic caching, and response filtering. AI middleware is a superset of API gateway functionality, designed for the non-deterministic nature of LLM calls.

Do I need AI middleware or can I just call the LLM API directly?

For a single use case with one model, direct API calls are simpler and faster to iterate on. You need middleware when routing between multiple models, needing observability in production, complying with data privacy rules, or managing token costs across a team. See our [AI agent observability guide](/blog/ai-agent-observability/) for what production monitoring looks like.

What is an LLM gateway?

An LLM gateway is a request-level proxy that sits between your application and one or more LLM APIs. It handles routing, caching, authentication, rate limiting, and cost tracking for individual model calls. It is the most common form of AI middleware, and is often what practitioners mean when they use the two terms interchangeably.

What are the best open-source AI middleware tools?

LiteLLM is the most widely adopted open-source LLM gateway — a single proxy supporting 100+ models with a unified API. LangChain provides agent middleware hooks (before_model, after_model) for orchestration-level control. LlamaIndex is best for RAG-heavy pipelines that need structured data retrieval.

How does AI middleware handle data privacy and security?

Mature AI middleware implements PII scrubbing before prompts reach external APIs, prompt injection detection, RBAC for model access, and immutable audit logs for compliance. For a full security checklist, see our [AI agent security guide](/blog/ai-agent-security/).

AI Middleware: The Glue Between LLMs and Business Systems

AI middleware connects LLMs to business systems with routing, security & caching. LEARN when you need it and which tools work best. Engineering guide.

Frequently Asked Questions

What is the difference between AI middleware and an API gateway?: An API gateway handles authentication, rate limiting, and routing for HTTP APIs. AI middleware does all of that plus LLM-specific concerns: model routing, prompt sanitization, token cost tracking, semantic caching, and response filtering. AI middleware is a superset of API gateway functionality, designed for the non-deterministic nature of LLM calls.
Do I need AI middleware or can I just call the LLM API directly?: For a single use case with one model, direct API calls are simpler and faster to iterate on. You need middleware when routing between multiple models, needing observability in production, complying with data privacy rules, or managing token costs across a team. See our [AI agent observability guide](/blog/ai-agent-observability/) for what production monitoring looks like.
What is an LLM gateway?: An LLM gateway is a request-level proxy that sits between your application and one or more LLM APIs. It handles routing, caching, authentication, rate limiting, and cost tracking for individual model calls. It is the most common form of AI middleware, and is often what practitioners mean when they use the two terms interchangeably.
What are the best open-source AI middleware tools?: LiteLLM is the most widely adopted open-source LLM gateway — a single proxy supporting 100+ models with a unified API. LangChain provides agent middleware hooks (before_model, after_model) for orchestration-level control. LlamaIndex is best for RAG-heavy pipelines that need structured data retrieval.
How does AI middleware handle data privacy and security?: Mature AI middleware implements PII scrubbing before prompts reach external APIs, prompt injection detection, RBAC for model access, and immutable audit logs for compliance. For a full security checklist, see our [AI agent security guide](/blog/ai-agent-security/).