Can you use Rivora from Claude or ChatGPT?

Yes. Rivora is in the Claude and ChatGPT directories: sign in with your Rivora account from Claude or ChatGPT, find a feature your team approved by its name, and read every approved part, its sources and its open questions before anyone builds. You think it through in your assistant; Rivora Connect keeps what your team confirmed.

What can you do with Rivora in Claude or ChatGPT today?

• Find a feature by name. Ask in plain words: "Find the Rivora plan for invoice reminders."
• Read the approved plan. The feature, its business rules and every approved part, as your team approved it.
• See where each part came from. The sources each part used.
• Check what is still open. The open questions on the feature, so you know what to ask before building.
• Send a suggestion back. If something in the approved plan looks wrong, the assistant can send a suggestion, if you allowed it when you signed in. A person reviews each one on the feature. An approved part changes only when someone chooses to change it.

The same connection works for coding agents. Approved work goes to coding agents such as Claude Code, Cursor and Codex, with the decisions and rules attached.

Why use Rivora alongside a Claude or ChatGPT project?

A project is a good place to think. It holds the files and instructions someone gave it, for the people they share it with.

Rivora holds what your whole team confirmed, feature after feature: decisions, rules and clarifications, each with where it applies and who confirmed it. Connect Rivora, and your assistant reads each feature's approved plan, with the business rules and sources your team confirmed. Keep using your projects alongside it. (See the comparison: Claude Projects.)

How do you connect it?

1. Find Rivora in the directory. In Claude, open the connectors directory. In ChatGPT, open Settings, then Plugins. Search for "Rivora".
2. Sign in with your Rivora account. Pick the workspace if you have more than one. No token is copied.
3. Ask for a feature by name.

For Claude Code, install the Rivora plugin and sign in from /mcp. Codex and Cursor connect the same way: sign in with Rivora from the agent.

What does the assistant see?

• Approved plans in the workspace you chose. Drafts never leave Rivora.
• Only what your team approved: a person approves each part, sends it back with a comment, or edits it.
• You can see and end each connection in Rivora, under Settings, Integrations, Agent access.

We don't use your content to train general-purpose AI models, and we will not knowingly permit our AI providers to use it for that. What Rivora learns stays in your workspace, as your product's memory.

Can my assistant ask Rivora questions about our product?

Not today. Today your assistant reads approved feature plans. We're building memory for your AI assistant: short answers from your team's approved memory, with sources, from inside Claude or ChatGPT. It is in development. Looking for this? Talk to us.

Questions

Do I need a Rivora account?

Yes. Rivora is in early access, and Rivora Connect sets up every team directly. Start with the free 30-minute session.

Is the connector made by Anthropic or OpenAI?

Rivora Connect makes it. Rivora is in the Claude and ChatGPT directories; in Claude's directory it is listed as a Community connector.

Can it change our approved plans?

No. If you allowed it when you signed in, it can send suggestions for a person to review. An approved part changes only when someone chooses to change it.