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 hierarchical and peer-to-peer agent systems?

Hierarchical systems use a tree-structured command chain where a supervisor orchestrates specialist and worker agents. Peer-to-peer systems have no central authority — agents communicate directly with each other and coordinate through shared state or message passing. See our [agent orchestration guide](/blog/ai-agent-orchestration/) for a full overview of patterns.

When should I use a hierarchical agent architecture?

Use hierarchical agents when you have 10+ agents to coordinate, need strong observability, or are tackling enterprise tasks where a clear chain of accountability matters. Hierarchical patterns contain error amplification to 4.4× vs 17.2× for independent agents, according to Google Research.

What is a supervisor agent in a multi-agent system?

A supervisor agent decomposes a user goal into subtasks, delegates them to worker agents, monitors progress, and synthesizes the final result. Critically, supervisors orchestrate and evaluate — they do not generate the final answer themselves.

What is the fan-out pattern in agent orchestration?

Fan-out is a centralized orchestration pattern where one orchestrator dispatches multiple subtasks to worker agents in parallel, then collects and synthesizes the results (fan-in). It achieves 80.9% performance improvement on parallelizable tasks vs sequential execution, per Google Research.

Can you mix hierarchical and peer-to-peer agents?

Yes — hybrid architectures are the production default. A common pattern is centralized planning (hierarchical supervisor decomposes the goal) with decentralized execution (worker agents coordinate peer-to-peer for parallel subtasks). Most enterprise systems built with [LangGraph, CrewAI, or AutoGen](/blog/ag2-vs-crewai-vs-langgraph-openai-agents-sdk/) use this approach.

Hierarchical vs Peer-to-Peer Agents: Which to Use

COMPLETE guide to hierarchical vs peer-to-peer agents. Compare control, scalability, fault tolerance & real-world use cases. Choose the RIGHT pattern for your system.

Frequently Asked Questions

What is the difference between hierarchical and peer-to-peer agent systems?: Hierarchical systems use a tree-structured command chain where a supervisor orchestrates specialist and worker agents. Peer-to-peer systems have no central authority — agents communicate directly with each other and coordinate through shared state or message passing. See our [agent orchestration guide](/blog/ai-agent-orchestration/) for a full overview of patterns.
When should I use a hierarchical agent architecture?: Use hierarchical agents when you have 10+ agents to coordinate, need strong observability, or are tackling enterprise tasks where a clear chain of accountability matters. Hierarchical patterns contain error amplification to 4.4× vs 17.2× for independent agents, according to Google Research.
What is a supervisor agent in a multi-agent system?: A supervisor agent decomposes a user goal into subtasks, delegates them to worker agents, monitors progress, and synthesizes the final result. Critically, supervisors orchestrate and evaluate — they do not generate the final answer themselves.
What is the fan-out pattern in agent orchestration?: Fan-out is a centralized orchestration pattern where one orchestrator dispatches multiple subtasks to worker agents in parallel, then collects and synthesizes the results (fan-in). It achieves 80.9% performance improvement on parallelizable tasks vs sequential execution, per Google Research.
Can you mix hierarchical and peer-to-peer agents?: Yes — hybrid architectures are the production default. A common pattern is centralized planning (hierarchical supervisor decomposes the goal) with decentralized execution (worker agents coordinate peer-to-peer for parallel subtasks). Most enterprise systems built with [LangGraph, CrewAI, or AutoGen](/blog/ag2-vs-crewai-vs-langgraph-openai-agents-sdk/) use this approach.