I’m having some pretty good luck having Codex w/ GPT 5.5 draw images of really good boards and charts and then recreate them in Lucid via MCP with a visual feedback loop to iterate to get the board to get pretty close to the image. However I’m running into issues doing fine grain editing such as change this font to a certain size. The coding agent and I are unaware of what endpoints are available and how they all work, and we need to build up a skill for what we can and can’t do via the MCP. We also should do a gap analysis for what capabilities could be added to the MCP down the road.
Idea
Agent Skill for Lucid MCP
+1
+12Hi
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.
For more information about how we manage feedback in this community, please take a look at this post:
Create an account in the community
A Lucid or airfocus account is required to interact with the Community, and your participation is subject to the Supplemental Lucid Community Terms. You may not participate in the Community if you are under 18. You will be redirected to the Lucid or airfocus app to log in.
Log in to the community
A Lucid or airfocus account is required to interact with the Community, and your participation is subject to the Supplemental Lucid Community Terms. You may not participate in the Community if you are under 18. You will be redirected to the Lucid or airfocus app to log in.
Log in with Lucid Log in with airfocus
or
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.
Your privacy choices
