Freeing Up Flow With Claude Code & Linear MCP

Coding with AI 5 min read

I’ve been experimenting with using Claude Code to create Linear issues instead of filling them out by hand, and it’s been a bigger workflow upgrade than I’d expected it to be. Too often I’ve made or come across Linear issues that are just a line or two, jotted down from someone’s memory and now turned inscrutable. Having Claude flesh out the details is a relief, especially when it then includes the Linear issue ID in its branch, so that the PR it produces gets automatically linked.

The Setup

In our repository’s CLAUDE.md file, I’ve added a section with the team and the project I’m currently working on:

## Linear Integration

All issues for web client work should be created on the **Day One Web (DOWEB)** team.

### IDs

- **Team:** Day One Web (DOWEB) - `<team-id>`
- **Project:** Daily Chat: Cross-Platform - `<project-id>`

### Labels

- **Platform → Web** - `<label-id>` - Use on all issues for the web client
- **Type → Bug** - `<label-id>` - Use for issues that are about bugs</label-id></label-id></project-id></team-id>

This way Claude doesn’t have to make a couple of MCP calls to get team and project IDs every time I start a new round of work. Claude Code already knows.

A Real Example

Here’s an actual prompt I used last week:

Please help me add two linear issues for the web in daily chat project:

  ☐ Regenerate response button should add a new ai chat message after the
    existing one instead of replacing the existing one. The existing ones
    should be passed in. The prompt should be updated to avoid asking similar
    questions in a row. If the last prompt was trying to dive deeper into a
    concept, the next response should be significantly different, perhaps
    exploring a different topic, or asking if the user has more events to
    write about.

  ☐ When returning to chat after N minutes of silence, the chat should sense
    their return and automatically generates a new AI prompt

  These can actually go into one issue. Part of the issue should also be
  updating the Documentation repo with this new behavior

Two things to notice:

  • Descriptive but concise. I said what needed to be done without over-explaining. A few good sentences, not a wall of text.
  • I changed my mind mid-prompt. I started asking for two issues, then realized they fit together as one. I didn’t have to stop and rewrite—Claude Code just rolled with it.

What It Created

From those few lines, Claude Code produced a well-structured issue with all the pieces I’d normally build by hand. So you don’t get bored with the details, here’s an abstracted view of what the issue looked like:

# [Title] Improve AI response regeneration and add idle return detection

## Summary
<brief description="" of="" what="" this="" issue="" covers="" and="" why="">

## Features

### 1. Regenerate Response Behavior Change
**Current behavior:** [what happens now]
**New behavior:**
- Bullet points describing the change
- Implementation details
- Edge cases to handle

### 2. Idle Return Detection
**New feature:** Description of the new capability
- Detection logic
- Response behavior

## Tasks
- [ ] Task 1..
- [ ] Task 2..
- [ ] Task 3..
- [ ] Update Documentation repo

## Acceptance Criteria
- Criterion 1
- Criterion 2
- Criterion 3</brief>

It structured features with current vs. new behavior, broke down tasks into a checklist, and added acceptance criteria. All from a casual prompt.

Why This Works

Linear’s UI is fine, but there’s mental friction in filling out fields, formatting markdown, and organizing thoughts into sections. With Claude Code, I just describe what I need as it’s pouring out of my brain, and it handles the structure. Leaving me free to keep my energetic momentum & motivation up, which helps nurture a “flow state”. And may be especially helpful when dealing with ADHD, based on what I’ve read in Hyper Focus, by Jesse J. Anderson.

Leave a Comment

Your email address will not be published. Required fields are marked *