Ability to recall prompt used for AI flowcharts

Hi ​@Chris_B, thank you for posting in the Lucid Community!

Would you mind elaborating on what you’re hoping to accomplish? Are you hoping to see a full chat history of previous prompts and chats within Lucid AI?

Unfortunately, a full Lucid AI chat history isn’t currently supported in Lucid. Starting a new chat deletes your previous chat. We’re very interested in your feedback and committed to continually improving our products. 

I’ve converted this to an Idea, which will be housed within the Product Feedback section of this community. We encourage anyone else who’s interested in this to upvote this post and share any additional details about your use case or what you’d like to see in this experience. 

Finally, for more information on how Lucid manages feedback via this community, take a look at this post:

Hello Ambar,

Thank you for reaching out and asking for clarification and additional explanation.

With regard to firms that work in the financial services space, the likelihood that firms will need to retain and reproduce AI prompts for regulatory review by FINRA*, the SEC, and state regulators is high and increasing. As AI becomes embedded in compliance, supervision, onboarding, and client‑facing workflows, regulators are shifting focus from outputs to governance, controls, and decision traceability.

The OECD AI Principles, adopted by the U.S. and widely referenced by global regulators, explicitly emphasize transparency, explainability, and accountability across the AI lifecycle. These principles require that organizations be able to explain how an AI system reached a result, not merely what it produced. In practice, prompts are a core control mechanism that materially influence AI behavior and outcomes, making them part of the system’s decision logic rather than informal artifacts.

Similarly, the EU AI Act reinforces this direction through formal record‑keeping, traceability, and technical documentation requirements. While the Act does not use the word “prompt” explicitly, its obligations around reproducibility, audit trails, and post‑market monitoring logically extend to prompt design, versioning, and changes—particularly for systems influencing regulated decisions. These frameworks collectively reject opaque “black‑box” AI in favor of systems that can be examined, challenged, and corrected.

From a compliance and risk‑management perspective, the ability to retrieve prior prompts enables:

• Demonstrable transparency during regulatory exams
• Explainability of AI‑assisted decisions and workflows
• Controlled updates and documented evolution of process mappings

Prompt retention therefore becomes not just a technical convenience, but a defensible governance control aligned with emerging regulatory expectations across U.S. and global regimes.

I welcome further conversation on this topic if deemed necessary. 

 *Artificial Intelligence (AI) | FINRA.org

​​ ​​​​​*Regulatory Notice 24-09 | FINRA.org

** The thoughts shared are mine alone and do not represent any firm, entity or regulatory entity. 

What I am doing at the moment is copying the full prompt history into the note on the document page.
This is handy not only for the documentation as you mention, but also for reusing some of the prompts when I need similar results/structure.

But overall I do agree that full-fledged history tab with document attribution to the conversation could be very useful too.