> ## Documentation Index > Fetch the complete documentation index at: https://docs.corunner.ai/llms.txt > Use this file to discover all available pages before exploring further. # Multimodal Reasoning: Screenshots, Diagrams, and Docs > Drop screenshots, architecture diagrams, whiteboard photos, product mockups, or documents into any thread. Corunner reads them and proposes the next action. Corunner reads images and documents directly inside chat. You do not have to describe what is on the screen. Post a screenshot, a diagram, a whiteboard photo, a mockup, or a document into any thread and mention `@Corunner`. Corunner identifies what is in the image, correlates it with your connected tools, and proposes the next action right in the conversation. ## What you can drop in * Architecture and system diagrams * Product screenshots and error dialogs * Whiteboard photos * Design mockups and Figma screenshots * Documents (PDFs and images of pages) ## Example prompts ```text Architecture diagram theme={null} @Corunner what's the risk in this system? ``` ```text Error screenshot theme={null} @Corunner has anyone seen this error before? Pull related incidents. ``` ```text Product mockup theme={null} @Corunner file a Linear issue for the empty-state redesign, attach the mockup. ``` ## What Corunner does with the image 1. Identifies the components, text, or artifacts in the image. 2. Maps them to your organizational memory: repositories, owners, past decisions, related incidents. 3. Proposes an action (file an issue, DM the owner, summarize the mockup for a review) with the image attached. 4. If the proposed action needs approval, the message shows **Approve**, **Edit**, and **Reject** controls. ## Related How Corunner runs inside Slack and Teams. Act on proposed actions without leaving the thread.