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 are agentic workflows?

An agentic workflow is a process in which an AI system decides its own next step at runtime instead of following a hardcoded sequence. The model reads the current state, chooses a tool or sub-task, evaluates the result, and loops until the goal is reached or a stop condition fires. Planning, tool use, memory, and a feedback loop are what make a workflow agentic.

What is the difference between an agentic workflow and an AI agent?

An AI agent is the actor — a model with tools, memory, and a reasoning loop. An agentic workflow is the orchestration layer around it: how the goal is framed, how state moves between steps, which agent handles which stage, how errors are retried, and when to stop or escalate to a human. One agent can run inside many workflows, and one workflow can coordinate several agents. See our guide to [AI agent orchestration](/blog/ai-agent-orchestration/) for how the layers fit.

What are the main agentic workflow patterns?

Seven patterns cover almost everything shipping in production: prompt chaining, routing, parallelization, orchestrator–workers, evaluator–optimizer (reflection), the ReAct tool-use loop, and the fully autonomous planner loop. The first five run on predefined code paths; the last two let the model direct itself.

When should you not use an agentic workflow?

Skip the agentic layer when the task has a fixed, known path, when latency budgets are under a second, when every decision needs a deterministic audit trail, or when a wrong action is irreversible and unguarded. A single LLM call with good retrieval solves more problems than most teams expect — Anthropic's engineering team recommends starting there and adding autonomy only when it measurably helps.

How much do agentic workflows cost to run?

Cost scales with the number of model calls, not the number of tasks. A simple chain of four steps costs roughly four times a single call; an evaluator–optimizer loop with three revision rounds can cost six to eight times; an open-ended agent loop has no ceiling until you set one. Always cap iterations and token spend per run — see our breakdown of [AI agent cost](/blog/ai-agent-cost/) for the full token math.

Agentic Workflows Explained: Patterns, Examples & Use Cases

Agentic workflows explained: the 7 core patterns, real examples, cost math, and when NOT to use them. Build your first workflow today.

Frequently Asked Questions

What are agentic workflows?: An agentic workflow is a process in which an AI system decides its own next step at runtime instead of following a hardcoded sequence. The model reads the current state, chooses a tool or sub-task, evaluates the result, and loops until the goal is reached or a stop condition fires. Planning, tool use, memory, and a feedback loop are what make a workflow agentic.
What is the difference between an agentic workflow and an AI agent?: An AI agent is the actor — a model with tools, memory, and a reasoning loop. An agentic workflow is the orchestration layer around it: how the goal is framed, how state moves between steps, which agent handles which stage, how errors are retried, and when to stop or escalate to a human. One agent can run inside many workflows, and one workflow can coordinate several agents. See our guide to [AI agent orchestration](/blog/ai-agent-orchestration/) for how the layers fit.
What are the main agentic workflow patterns?: Seven patterns cover almost everything shipping in production: prompt chaining, routing, parallelization, orchestrator–workers, evaluator–optimizer (reflection), the ReAct tool-use loop, and the fully autonomous planner loop. The first five run on predefined code paths; the last two let the model direct itself.
When should you not use an agentic workflow?: Skip the agentic layer when the task has a fixed, known path, when latency budgets are under a second, when every decision needs a deterministic audit trail, or when a wrong action is irreversible and unguarded. A single LLM call with good retrieval solves more problems than most teams expect — Anthropic's engineering team recommends starting there and adding autonomy only when it measurably helps.
How much do agentic workflows cost to run?: Cost scales with the number of model calls, not the number of tasks. A simple chain of four steps costs roughly four times a single call; an evaluator–optimizer loop with three revision rounds can cost six to eight times; an open-ended agent loop has no ceiling until you set one. Always cap iterations and token spend per run — see our breakdown of [AI agent cost](/blog/ai-agent-cost/) for the full token math.