Draft and post experiment summaries from Jupyter to Linear
Read a Jupyter notebook export from Google Drive and create a structured Linear issue with results
Overview
Wrapping up an experiment means writing up what you tried, what worked, and whats next—and that write-up often gets delayed or skipped. This template reads your notebook output and creates a Linear issue with the key sections already filled in, so your team can act on results faster.
How it works
- Reads the exported Jupyter notebook output from Google Drive
- Identifies the hypothesis, approach, key results, and next steps
- Creates a new Linear issue with a structured description
- Assigns the issue to the specified team member
- Sets priority to high if results show more than 10% improvement, otherwise medium
Who this is for
Data scientists and ML engineers who want experiment results documented and actionable without spending time on manual write-ups.
Suggested prompt
Read the exported output from my Jupyter notebook stored in Google Drive called [1. Notebook export name]. Identify the experiment hypothesis, the approach taken, the key results, and any open questions or recommended next steps. Then create a new issue in my Linear project called [2. Project name] with the title "[3. Experiment name] Results" and populate the description with those four sections. Assign the issue to [4. Team member name] and set the priority to medium unless the results include a performance improvement of more than 10%, in which case set it to high.
Frequently asked questions
Can I change the priority threshold?
Yes, adjust 10% in the prompt to any threshold that fits your teams standards.
What if my notebook is in a different format?
Modify the prompt to match your export structure, whether its HTML, PDF, or markdown.
Can I post to Jira instead of Linear?
Yes, swap Linear for Jira in the prompt and connect Jira to your MCP server.
Can I add custom fields to the issue?
Yes, modify the prompt to include labels, milestones, or any other fields your Linear setup uses.