Automating my morning with Linear MCP Server and Claude

Every morning, I'd sit down with a cup of Yorkshire tea (only the best) and do the same routine:

- Open Linear

- Listen to my daily Pulse on the teams project updates

- Next scan through my assigned tickets

- Check which ones were urgent or had new comments

- Plan my day based on priorities and deadlines

I decided to automate this entire process using Claude AI and Linear's new MCP (Model Context Protocol) server. Now, every weekday at 7 AM, I get an email with:

- Prioritised ticket analysis with dependency mapping

- Comment analysis showing what's waiting on others vs. ready to work

- Recommended daily plan with time estimates and logical ordering

- Actionable next steps for each ticket

Here's what an 'example' typical morning email looks like:

How it works

To get this working there are three key components:

1. Linear MCP Server Integration

Linear's MCP server provides direct access to my tickets, comments, and project data. Instead of scraping or using REST APIs, the MCP connection gives Claude deep context about my work.

2. Claude AI Analysis

A carefully crafted prompt analyses my tickets for:

- Priority levels and due dates

- Comment patterns indicating waiting states

- Dependencies between tickets

- Estimate complexity for daily planning

3. Automated Scheduling

Python script runs daily via cron/launchd, connects to Claude API, analyses via MCP, and emails results.

## The Implementation

Here's the core of how it works including the system prompt:

```python

def get_linear_analysis():

   """Use Claude API with Linear MCP to analyze tickets"""

   client = anthropic.Anthropic(api_key=CLAUDE_API_KEY)

   prompt = """Using the Linear MCP server, analyze my assigned tickets and help me create a plan for today.

Please provide:

1. **High Priority & Urgent Items** - What needs immediate attention

2. **Dependencies & Blockers** - What's waiting on others or blocking progress  

3. **Recommended Daily Plan** - Prioritized order with time estimates

For each ticket, include:

- Any relevant comments that show waiting states or recent updates

- Specific next actions to start the work

- Dependencies on other tickets

Focus on actionable insights and realistic daily planning."""

   response = client.messages.create(

       model="claude-3-sonnet-20241022",

       max_tokens=4000,

       messages=[{"role": "user", "content": prompt}]

   )

   return response.content[0].text

```

What I learnt

1. Context is Everything

The Linear MCP connection gives Claude access to comment history, project context, and team dynamics. This isn't just parsing ticket titles - it's understanding the full story.

2. Prompting for Action, Not Information

Instead of asking "what are my tickets?", I ask "what should I do today?" The AI provides decision-making, not just data retrieval.

3. Consistent Execution Beats Perfect Analysis

A good plan executed daily beats a perfect plan executed occasionally. Automation ensures consistency.

Follow for more!