Describe the picture you want. Your AI makes a few versions in RobinReach, looks at them, attaches the best one to your post and keeps the others in your media library.
Make an image for Friday's Instagram post: our speckled mugs on a linen tablecloth, soft morning light, portrait. Try a few and attach the best one.
I made four versions and attached the one with the softest light to Friday's post. The other three are in your media library if you'd rather swap. That used 4 of your 50 AI images this month.
Speckled stoneware mugs on a linen tablecloth, soft morning light, portrait
An example conversation with an invented brand. Photos from Unsplash stand in for AI images.
Your assistant brings the thinking and the writing. RobinReach gives it hands: these are the parts of RobinReach it works with to get this done.
Claude, ChatGPT, Gemini, Codex or Cursor. You brief it the way you would brief a person, and it decides which RobinReach actions to take, in what order.
Connect yours in two minutesMakes images from your description in the shape and style you ask for, saves every version to your media library and attaches the one you want to a post.
Learn moreWhat happens between your message and the result. The small labels are the RobinReach actions your AI calls, in case you are curious.
Add RobinReach to Claude or ChatGPT as a connector with one link, click Connect and sign in to RobinReach. There is no API key to copy. Other assistants use the link with your API key, from Integrations in RobinReach. It takes about two minutes and you never repeat it.
Your plan includes a monthly allowance of AI images. Your AI checks what's left before it starts.
Each one is saved to the RobinPixel folder of your media library. Images come in five shapes and six styles, made with OpenAI's GPT Image or Google's Gemini.
Your AI sees the images themselves, so it can choose the one that matches what you asked for.
The chosen image goes straight onto the post. The other three wait in your library.
Everything your AI did is in RobinReach: the calendar, the posts and their status. Edit, move or delete anything from there.
Paste it into your assistant and change the details to fit your brand.
Make an image for Friday's Instagram post: our speckled mugs on a linen tablecloth, soft morning light, portrait. Try a few and attach the best one.
Copy one, paste it into Claude or ChatGPT, and swap in your own details.
Make a square image for tomorrow's post about our autumn menu: warm colours, wooden table, no text.
Try the same image with Google's Gemini and show me both.
Make a 16:9 image for our LinkedIn post about the new studio, vintage style, in our brand colours.
Replace the image on Saturday's post with a new one: same idea, brighter.
How many AI images do we have left this month?
Every use case is a recipe of RobinReach features. These share some of this one's.
Your AI looks at each photo in a folder and writes a caption and alt text that match it.
+ Media library 1 feature See how it works EssentialYour AI studies your past posts, remembers your corrections and writes like you on every network.
+ Brand voice 1 feature See how it works EssentialThis week against last week, your best posts and a clear next step, in plain words.
+ Analytics 1 feature See how it worksConnect Claude or ChatGPT to RobinReach once and ask for the picture you need. Included in Bloom, Thrive and Scale.