Obsidian Email Chain Processor

Extract participants from email chains, create People profiles, and generate summaries from email thread content.

Quick Start

"Process the email" or "Process email chain" = Complete workflow:

1. •Extract participants from email chain (From headers only, not To/CC)
2. •Create missing People profiles
3. •Update note with linked participant list
4. •Generate summary from email chain content and insert into Summary section

This is the default behavior - all processing steps are performed automatically unless you request a specific step only.

Capabilities

1. •Extract Participants: Parse email From headers to identify senders
2. •Create People Profiles: Generate linked profiles for email participants
3. •Generate Summaries: Analyze email chain and create structured summary

File Structure

Email chain notes use format: YYYY-MM-DD - topic name.md in ~/Documents/Obsidian/HPE/Notes/

Expected Structure

---
when: YYYY-MM-DD
tags:
  - note
---
# Participants


# Summary



# Email Chain

[Raw email content here]

Finding Email Notes

# Today's email notes
find ~/Documents/Obsidian/HPE/Notes -name "$(date +%Y-%m-%d)*.md"

# Search for files with "Email Chain" section
find ~/Documents/Obsidian/HPE/Notes -type f -name "*.md" -exec grep -l "# Email Chain" {} \;

# Specific email note by name
find ~/Documents/Obsidian/HPE/Notes -iname "*ops*ramp*.md"

Processing Workflow

Step 1: Extract Participants from Email Chain

Parse the email chain to extract unique senders:

1. •Look for From: or **From:** headers in email chain
2. •Extract name in "Last, First" format (e.g., "Vobbilisetty, Suresh")
3. •Build list of unique senders (ignore To/CC lists)
4. •Maintain chronological order (first sender first)

Email header patterns to match:

From: Last, First <email@hpe.com>
**From:** Last, First <email@hpe.com>

Step 2: Create/Update People Profiles

For each unique sender:

1. •Check if profile exists: ~/Documents/Obsidian/HPE/People/Last, First.md
2. •If missing, create new profile:
   markdown

   ---
   tags:
     - person
   ---
   
   # Last, First
   
   ## Contact
   - Email: email@hpe.com
   
   ## Aliases
   - [[Last, First|First Last]]
   

3. •Extract email address from From header
4. •If profile exists but email missing, add it to Contact section

Step 3: Update Participants Section

Replace empty # Participants section with linked list:

# Participants

- [[Vobbilisetty, Suresh|Suresh Vobbilisetty]]
- [[Sadananda, Ravi Kiran Srirangam|Ravi Kiran Sadananda]]
- [[Daniel, Binu|Binu Daniel]]
- [[Vanteru, Bhanu|Bhanu Vanteru]]

Format rules:

• •Link format: [[Last, First|First Last]]
• •Listed in order of first appearance in email chain
• •One participant per line with bullet point

Step 4: Generate Detailed AI Summary (Automatic)

The script automatically generates comprehensive, detailed summaries using AI analysis:

Detailed AI Summary includes:

• •✅ Detailed multi-paragraph overview with full context
• •✅ Technical issues broken down by root cause with specific details
• •✅ Stakeholder mapping organized by role
• •✅ Action items organized by timeline (Immediate, Short-Term, Long-Term) with owners
• •✅ Process breakdown analysis and communication gaps
• •✅ Tables for ticket/issue tracking
• •✅ Timeline of events with timestamps
• •✅ Specific technical details (serial numbers, configurations, error messages)
• •✅ Customer impact assessment and frustration quotes
• •✅ Root cause summary (technical + process)
• •✅ Next steps prioritized

Automatic Analysis Process:

1. •Script analyzes up to 30KB of email content
2. •AI generates comprehensive summary (30-60 second processing time)
3. •Falls back to enhanced basic summary only if AI is unavailable

Example Detailed Summary Structure:

## Overview
[2-3 detailed paragraphs with customer, timeline, urgency, context]

## Technical Issues Identified
### Issue #1: [Specific Problem Name]
**Problem:** [Detailed description]
**Root Cause:** [Technical analysis]
**Solution Required:** [Fix plan with timeline]

### Affected Tickets Table
| Ticket | Summary | Status | Root Cause |

## Process & Escalation Issues
- Communication breakdowns identified
- Leadership concerns documented

## Action Items & Responsibilities
### Immediate Actions (Due: [Date])
- [x] Completed item with owner
- [ ] Pending item with owner and deadline

## Key Stakeholders
### Engineering Leadership
- Name - Specific role and involvement

## Root Cause Summary
### Technical Failures
### Process Failures  
### Customer Impact Duration

## Next Steps
[Prioritized action list]

Enhanced Basic Summary (Fallback only): If AI is unavailable, generates enhanced basic summary with:

• •Extracted ticket references
• •Key dates mentioned
• •Urgency indicators
• •Customer/project names from subject

When to Use

• •User requests "process email", "extract participants", or "summarize email"
• •Note file contains # Email Chain section with email content
• •After creating note from copied email thread

Script Usage

# Process email chain note
python ~/.copilot/skills/obsidian-email/scripts/process_email.py "email-note-file.md"

The script automatically:

1. •Parses email headers to extract unique senders
2. •Creates missing People profiles with email addresses
3. •Updates Participants section with linked names
4. •Generates comprehensive summary from email content
5. •Updates Summary section in the note

Example output:

✓ Extracted 5 unique participants
  - Vobbilisetty, Suresh (suresh.vobbilisetty@hpe.com)
  - Sadananda, Ravi Kiran Srirangam (ravikiransrirangam.sadananda@hpe.com)
  - Daniel, Binu (binu.daniel@hpe.com)
  - Vanteru, Bhanu (bhanu.vanteru@hpe.com)
  - Yun, Stella (xiaoyang.yun@hpe.com)
✓ Created 2 new People profiles
✓ Updated Participants section
✓ Generated summary (4 sections)
✓ Updated email note

Email Header Formats

The script handles multiple email header formats:

Format 1: Plain text with colon

From: Last, First <email@hpe.com>
Date: Friday, January 23, 2026 at 7:43 AM
To: Person1 <email1@hpe.com>, Person2 <email2@hpe.com>

Format 2: Bold markdown

**From:** Last, First <email@hpe.com>
**Sent:** Friday, January 23, 2026 5:42 AM
**To:** Person1 <email1@hpe.com>

Format 3: Mixed (Outlook style)

From: Last, First <email@hpe.com>
Date: Friday, January 23, 2026 at 7:43 AM

---

**From:** Last, First <email@hpe.com>
**Sent:** Friday, January 23, 2026 12:48 PM

Name Format Handling

The script handles various name formats:

• •Standard: Last, First → [[Last, First|First Last]]
• •Middle names: Last, First Middle → [[Last, First Middle|First Middle Last]]
• •Multi-part last names: Last Name, First → [[Last Name, First|First Last Name]]

Email Address Extraction

Extracts email from patterns:

• •<email@domain.com> - Standard format
• •[email@domain.com](mailto:...) - Markdown link format
• •email@domain.com - Plain text format

Integration with People Profiles

The script maintains consistency with existing People profiles:

1. •Check existing profiles before creating new ones
2. •Add email to Contact section if profile exists but email missing
3. •Use same formatting as obsidian-meeting skill for consistency
4. •Create aliases section for proper Obsidian linking

Example Workflow

# 1. Find the email note
find ~/Documents/Obsidian/HPE/Notes -iname "*ops*ramp*email*.md"

# 2. Process it
python ~/.copilot/skills/obsidian-email/scripts/process_email.py \
  "~/Documents/Obsidian/HPE/Notes/2026-01-23 - Ops Ramp Issues Email.md"

# Output shows progress and results

Common Issues

"No email chain found"

• •Ensure the note has a # Email Chain section
• •Check that email content is below the header

"No senders extracted"

• •Verify email has From: headers
• •Check email format matches supported patterns

"Failed to parse name"

• •Name should be in "Last, First" format in From header
• •Script will skip invalid formats and continue

Related Skills

• •obsidian-meeting: Process meeting transcripts with similar participant extraction
• •Both skills maintain consistent People profile format and linking conventions