Clear answers for your next step

Getting started

  • Can I use CogineWork as an individual?

    Yes. Run CogineWork on your computer, connect a supported AI service or local model, and use it for research, creation and development.

  • What makes a good first task?

    Start with a focused question, provide the necessary context, and say what you want to inspect. For example, compare a few public sources and organize their differences and open questions.

  • Which computers can I install it on?

    Downloads are available for macOS on Apple Silicon and Windows x64. Check the download page for current releases, system requirements and verification details.

  • Can a team use CogineWork too?

    Alongside self-managed local use, CogineWork supports organization-managed single-machine deployments. Team use should be planned around your actual deployment, configuration and ways of working.

Choosing your AI

  • Do I have to use one particular AI service?

    No. Choose a provider supported by the client and use your own account, API key or local model. Available models and capabilities vary by connection.

  • Can I use my existing Codex subscription?

    Authorize a ChatGPT account with the relevant Codex access. The Codex connection uses account authorization, not a standard OpenAI API key.

  • How do I connect OpenCode Go?

    Choose OpenCode Go in the client and configure the corresponding Go API key. Go and OpenCode Zen are separate connections; availability and charges follow your service plan.

  • How is model usage billed?

    Model use follows the billing rules of your connected account or provider. Connecting a subscription does not move its allowance to another provider. Local models also depend on your hardware and runtime.

  • Can I use local models?

    Supported local runtimes include Ollama and LM Studio. Start the runtime, then configure a model using the connection supported by the client.

Data and boundaries

  • Does local-first mean no data leaves my computer?

    Not always. Cloud models and external tools may receive the content needed for a task. The data path depends on your chosen models, tools and configuration.

Work and automation

  • What are Skills and MCP for?

    Skills organize reusable working methods. MCP connects configured tools and resources. Available actions depend on installed capabilities and the current configuration.

  • Do automations need my computer to stay on?

    Automations depend on their host environment staying available. A local scheduled task is not a cloud service that keeps running while your computer is off.