---
title: "Interact with agent-created visualizations in canvases"
description: "Cursor introduces canvases, allowing agents to represent information through interactive visual interfaces instead of chat text or markdown. In the Agents Window, these durable artifacts sit alongside tools including the terminal, browser, and source control, and agents can build dashboards, PR review interfaces, architecture diagrams, and eval-failure investigations. Canvases use a React-based UI library with first-party tables, boxes, diagrams, and charts, plus existing Cursor components such as diffs and to-do lists; skills can teach agents to create specific canvas types. Cursor reports that combining observability data from Datadog, Databricks, and Sentry MCPs with local debug files makes time-series analysis easier, while eval skills group rollout failures to expose hidden harness bugs. The company is also using canvases to visualize autoresearch progress and current hypotheses during client-performance experiments."
---

# Interact with agent-created visualizations in canvases

[Cursor](https://yomu.fyi/company/cursor) · Alex Vandak Maloney · Apr 15, 2026

**Type:** Announcement

## Summary

Cursor introduces canvases, allowing agents to represent information through interactive visual interfaces instead of chat text or markdown. In the Agents Window, these durable artifacts sit alongside tools including the terminal, browser, and source control, and agents can build dashboards, PR review interfaces, architecture diagrams, and eval-failure investigations. Canvases use a React-based UI library with first-party tables, boxes, diagrams, and charts, plus existing Cursor components such as diffs and to-do lists; skills can teach agents to create specific canvas types. Cursor reports that combining observability data from Datadog, Databricks, and Sentry MCPs with local debug files makes time-series analysis easier, while eval skills group rollout failures to expose hidden harness bugs. The company is also using canvases to visualize autoresearch progress and current hypotheses during client-performance experiments.

## Context

Plain-text chat and markdown tables can make information, especially time-series data and large diffs, difficult to interpret. Cursor describes canvases as part of an effort to increase information bandwidth and reduce friction in human-agent collaboration.

## Approach / What changed

Cursor renders canvases with a React-based UI library, first-party components, and existing tools such as diffs and to-do lists. Agents can create dashboards and other interactive interfaces, while skills teach them to generate specialized canvases for architecture diagrams, eval analysis, and other workflows.

## Takeaways

- Canvases are durable artifacts in Cursor’s Agents Window and can combine data from sources such as Datadog, Databricks, Sentry MCPs, and local debug files in a single visualization.
- A Cursor skill can read all rollouts in an eval, group failures, and build a canvas for investigating failure clusters and previously hidden harness bugs.
- Cursor uses canvases during autoresearch experiments so users can monitor progress and see the hypothesis currently being tested while optimization work runs.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [Data Visualization](https://yomu.fyi/topic/data-visualization), [Observability](https://yomu.fyi/topic/observability), [React](https://yomu.fyi/topic/react)

- Source: [Cursor](https://cursor.com/blog/canvas)
- Source URL: https://cursor.com/blog/canvas
- Ingested by Yomu: 2026-09-02T11:43:03.627Z

[Read original post](https://cursor.com/blog/canvas)