Build a benchmark image collage in Canva
The author had ChatGPT Agent log into Canva and build a collage of benchmark results; after over 75 minutes the result was poor as it fumbled the interface and drew shapes in Python. Task: Suggested implementation: Use ChatGPT Agent to extract benchmark images from OpenAI's announcement post and arrange them in Canva with red rectangles highlighting specific columns. The workflow requires the human to manually solve CAPTCHAs and log into Canva, then redirect the agent if it attempts to draw shapes in Python instead of using the Canva interface. Initial attempts may ignore Canva, requiring stricter prompts to enforce tool usage.; tools: ChatGPT Agent. Human role: The original post does not say how a person stayed involved.; limitations: Prerequisites include valid Canva credentials and the ability to manually bypass CAPTCHAs. Constraints involve the agent's tendency to default to Python for shape drawing rather than using the Canva UI, leading to rough results and extended processing times (over 75 minutes in reported cases). Simpler tasks like cover images may yield better performance. Source reliability is low due to HTTP 403 errors and truncated text..
Tools
Published on Agent Work Atlas:
Overview
The author had ChatGPT Agent log into Canva and build a collage of benchmark results; after over 75 minutes the result was poor as it fumbled the interface and drew shapes in Python.
Workflow
Suggested implementation: Use ChatGPT Agent to extract benchmark images from OpenAI's announcement post and arrange them in Canva with red rectangles highlighting specific columns. The workflow requires the human to manually solve CAPTCHAs and log into Canva, then redirect the agent if it attempts to draw shapes in Python instead of using the Canva interface. Initial attempts may ignore Canva, requiring stricter prompts to enforce tool usage.
Reported result
Outcomes are unknown as the source data is incomplete. Proposed measurable acceptance check: Verify that the final collage is created within Canva (not via Python scripts), that images are correctly aligned, and that red rectangles are drawn using Canva's native tools. Track time-to-completion and visual accuracy against a manual baseline.
Human role
The original post does not say how a person stayed involved.
Limitations
Prerequisites include valid Canva credentials and the ability to manually bypass CAPTCHAs. Constraints involve the agent's tendency to default to Python for shape drawing rather than using the Canva UI, leading to rough results and extended processing times (over 75 minutes in reported cases). Simpler tasks like cover images may yield better performance. Source reliability is low due to HTTP 403 errors and truncated text.
Try it
Act as a content designer using ChatGPT Agent. Task: Create a benchmark image collage in Canva. Inputs: [URL of OpenAI announcement post], [Canva login credentials], [specific columns to highlight]. Steps: 1. Instruct the agent to extract benchmark images from the provided URL. 2. Direct the agent to open Canva and arrange these images. 3. Specify that red rectangles must be drawn using Canva's interface, not Python. 4. If the agent fails to use Canva or gets stuck on CAPTCHA, pause and provide manual login assistance. 5. Review the output for alignment and correct tool usage. Deliverable: A Canva link containing the collage. Success Metric: Images are aligned, rectangles are native Canva shapes, and total time is under 60 minutes. Dry Run: Test with a single image first. Human Confirmation: Verify login status and shape creation method before finalizing.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.