Generate and edit images

OpenScience renders schematics, illustrations, graphical abstracts and image edits with its generate_image tool, and saves the result into your project. Ask for a clear visual and an output path, then inspect the saved image before using it.

Image routes

Image generation runs on one of three routes:

A personal OpenRouter key does not render images; connect the provider directly, or use Ace. Selecting a text-only or local chat model does not by itself provide image generation. The agent knows which route is active and says so when none is: it will not draw a diagram by hand with TikZ or SVG in its place.

Image-service usage has its own costs and terms. Verify the route before planning a workflow around generated images.

Generate a new illustration

Create a clean 16:9 conceptual illustration of the study workflow:
sample collection, measurement, analysis, and interpretation. Use short
labels, a white background, and colorblind-friendly accents. Save it as
results/study-workflow.png. This is a conceptual diagram, not measured data.

Specify composition, intended audience, labels, colors, background, and aspect ratio. Supported ratios are 1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16 and 21:9; ask for 2K when the figure will be printed.

Use a .png destination for the most straightforward compatible output (Ace and OpenAI also write .jpg and .webp). Ask for the exact saved path and open it in Files.

Method, pipeline and architecture diagrams go through the schematics skill. It describes the diagram component by component from your text, renders it with purpose: "schematic" (the tool prepends the publication framing adapted from K-Dense's scientific-schematics skill: white background, one sans-serif face, a colorblind-safe palette with one accent, one reading direction, your labels verbatim, nothing invented, no figure numbers inside the image), scores the render on five criteria against the threshold for the document type (a journal figure needs 8.5 of 10, slides 6.5), re-renders once from the critique when it falls short, and finishes at 2K. Conceptual figures and graphical abstracts use purpose: "illustration"; edits of an existing image use purpose: "edit".

Edit an existing image

Edit figures/workflow-draft.png. Increase label contrast, change the
background to white, and keep the four stages and their order. Save a
new file at figures/workflow-revised.png so the draft remains available.

Provide an existing PNG, JPEG, WebP, or GIF image as the input. State what must remain and what should change. Style references (earlier figures, a published diagram) can ride along on a Gemini or OpenAI key; Ace takes one image per request.

Review the result

Click the thumbnail or Open image on a generated-image card to view the original attachment. It remains available if the workspace file is later renamed, edited, or removed. Use Files to inspect the current version on disk. Images too large to attach open from their saved path instead.

Check text spelling, labels, arrows, order, visual consistency, and whether a requested edit changed other details. Ask for a specific correction using the saved image as the next input.

For a scientific concept, verify that the illustration reflects the intended mechanism and does not imply unsupported evidence. Keep any required attribution or disclosure with the final material.

Use measured data for scientific plots

Use Figures and plotting code for charts of experimental or simulated results. Image generation is useful for explanatory illustrations; it does not replace data analysis or produce trustworthy numerical measurements.

Save approved images as Results, and keep captions and source material in the project.