CogentSpec

Bridge overview

Choose how you work with your AI.

The Bridge connects your AI conversation to your CogentSpec project. Choose the browser, split-screen or desktop experience that suits you.

Your AI’s web chat

Web Bridge (Free)

Use ChatGPT, Claude or Qwen web chat alongside the CogentSpec work area in your browser. Connect the conversation so you can discuss your idea while keeping the project’s requirements and next steps in view.

Best for
Working in your browser with your preferred AI.
You need
Your AI web account and its supported browser connection.

Qwen Web is in preview.

Conversation and project, side by side

Split Screen Bridge

Keep your AI conversation on one side of your screen and the CogentSpec work area on the other. Discuss the work, review the specification and follow the project without switching back and forth between windows.

Best for
Seeing the conversation and project together as you work.
Available for
ChatGPT and Claude desktop workflows.

Your AI’s desktop app

Desktop UI Bridge

Work from ChatGPT Desktop or Claude Desktop with CogentSpec connected to your conversation. This option uses the AI desktop interface, rather than its web chat.

Best for
Staying in the desktop AI app you already use.
To connect
Choose Desktop UI Bridge in the ChatGPT or Claude menu, then paste the supplied command into your desktop chat and send it.

ChatGPT uses $gptdesk. Claude uses $claudedesk.

Qwen on your own hardware Preview

Qwen Local Bridge

Use a browser-based CogentSpec work area with Qwen running through Ollama on your own hardware. The interface is web-based; the model runs locally. You do not need to download the CogentSpec Desktop app.

Best for
Using your own installed Qwen model instead of hosted AI chat.
You need
Ollama running, a supported Qwen model installed and a browser that can reach Ollama.

Connect to Ollama and choose your model. The work area’s help screen explains the $qwenlocal connection command.