Process Meetings

Process meetings that have been synced from Granola by the background automation. Updates person pages, extracts tasks, and organizes meeting notes.

Background Execution

This skill supports background execution. When invoked:

1. Acknowledge: "Processing [N] meetings in the background. I'll let you know when done."
2. Process all meetings
3. On completion, provide summary: "[N] meetings processed. [X] person pages updated. [Y] action items created."

How It Works

Meetings are synced automatically every 30 minutes by a background process. This command reads those synced files and:

• Creates/updates person and company pages
• Extracts action items to 03-Tasks/Tasks.md
• Links everything together

No terminal commands are shown - the heavy lifting happens in the background.

Arguments

• No arguments: Process all unprocessed meetings from the last 7 days
• today: Only process today's meetings
• "search term": Find meetings by title/attendee
• --people-only: Only update person/company pages (skip tasks)
• --no-todos: Create notes but don't extract tasks
• --setup: Install/check background automation

Pre-flight: Granola Check

Granola sync uses the official Granola public API. Desktop and mobile recordings both come through it once your Granola API key is connected. If GRANOLA_API_KEY isn't set (checked in the environment, then the .env file at the vault root), say: "Granola isn't connected yet — run /granola-setup to add your Granola API key (requires a Granola Business plan)." and continue with any meetings already synced.

Process

Step 1: Check Background Sync Status

First, check if background sync is set up:

# Check for state file (indicates sync has run)
ls .scripts/meeting-intel/processed-meetings.json

If state file exists: Background sync is working. Continue to Step 2.

If state file doesn't exist:

"Background meeting sync isn't set up yet. This runs automatically every 30 minutes so /process-meetings doesn't need terminal commands.

To set up (one-time, takes 30 seconds):

cd .scripts/meeting-intel && ./install-automation.sh

Copy

Or run /process-meetings --setup and I'll do it for you.

Requirements:

• A Granola Business plan, with your Granola API key connected via /granola-setup
• An LLM API key in .env (GEMINI_API_KEY, ANTHROPIC_API_KEY, or OPENAI_API_KEY)"

If user runs --setup:

cd .scripts/meeting-intel && ./install-automation.sh

Step 2: Find Synced Meetings

Read the processed meetings state:

const state = JSON.parse(fs.readFileSync('.scripts/meeting-intel/processed-meetings.json'));

List meeting files in 00-Inbox/Meetings/:

find 00-Inbox/Meetings -name "*.md" -mtime -7 | head -50

For each meeting file:

1. Read frontmatter to get granola_id, participants, company, date
2. Check if person/company pages need updating
3. Check if tasks need extracting (look for unchecked items in "For Me" section)

Report findings:

"Found X synced meetings from the last 7 days. Y need person page updates, Z have unextracted tasks."

Step 3: Update Person Pages

For each participant in synced meetings:

1. Load user profile for email domain:

   Read System/user-profile.yaml → get email_domain
   

   Copy

2. Classify as Internal/External:

   • If participant email domain matches user's domain → Internal
   • Otherwise → External

3. Look up the person with the Work MCP lookup_person tool.

   • If lookup returns ambiguous: true, do not create a page. Surface the possible matches to the user.
   • If a match exists, update that existing page.

4. If no match exists, call the Work MCP create_person tool:

   • Pass name, role when known, emails from the meeting's attendees block, and location from that attendee's location field.
   • Pass the meeting company and a short source note when available.

---
type: person
name: "{Name}"
role: null
company: "{company from meeting}"
company_page: null
emails: ["{lowercased email, if available}"]
aliases: []
location: {internal|external}
last_interaction: {meeting date}
---
# {Name}

## Notes

*Auto-created from meeting on {date}*

## Recent Interactions

<!-- dex:auto:recent-interactions -->
- [{Meeting Title}](00-Inbox/Meetings/{date}/{slug}.md) — {date}
<!-- /dex:auto -->

## Key Context
 ```

5. **If page exists, add meeting to Recent Interactions:**
- Read existing page
- Add new meeting link under "## Recent Interactions"
- Keep max 20 entries (remove oldest if needed)
- Update "Last Interaction" in frontmatter

### Step 4: Update Company Pages

For each unique external company domain:

1. **Check if company page exists:** `05-Areas/Companies/{Company}.md`

2. **If doesn't exist, create it:**
```markdown
---
type: company
name: "{Company Name}"
domains: ["{lowercased domain}"]
website: "{website, if known}"
status: "Prospect"
---
# {Company Name}

## Key Contacts

<!-- dex:auto:key-contacts -->
- [[05-Areas/People/External/{Person}|{Person}]]
<!-- /dex:auto -->

## Meeting History

<!-- dex:auto:meeting-history -->
- [{Meeting Title}](00-Inbox/Meetings/{date}/{slug}.md) — {date}
<!-- /dex:auto -->

## Notes

*Auto-created from meeting on {date}*

3. If exists, update:
   • Add any new contacts to "Key Contacts"
   • Add meeting to "Meeting History"

Step 4.5: Semantic Enrichment (if QMD available)

Check if semantic search is available by looking for qmd in PATH.

If available, enhance meeting processing with meaning-based intelligence:

1. Detect implicit commitments: For each meeting's discussion notes, search semantically:

   qmd query "we should circle back on..." --limit 3
   qmd query "let me think about..." --limit 3
   

   Copy

   Catch soft commitments that regex action-item extraction misses.

   • Examples: "we should probably revisit the pricing model" → implicit action item
   • "I need to noodle on the migration approach" → implicit commitment
   • "Let's reconnect after the board meeting" → implicit follow-up

2. Link meetings to projects: For the meeting topic, search:

   qmd query "meeting topic/title" --limit 3
   

   Copy

   against 04-Projects/ to auto-link the meeting to relevant projects that keyword matching would miss.

3. Enrich person context: For each new person encountered, search:

   qmd query "person name + company" --limit 3
   

   Copy

   Find if they've been mentioned in other meetings/notes, even if they weren't a direct participant.

Integration:

• Add implicit commitments to the action items list with a note: "(detected — not explicitly stated)"
• Add project links to meeting frontmatter
• Merge person context into newly-created person pages
• If QMD unavailable, skip silently — regex extraction still works

Step 5: Extract Tasks (unless --no-todos or --people-only)

For each meeting with unextracted tasks:

1. Find action items in the "## Action Items > ### For Me" section

2. For each unchecked item (- [ ]):

   • Extract task description
   • Get task ID (format: ^task-YYYYMMDD-XXX)
   • Read pillar from meeting frontmatter

3. Create task using Work MCP:

   create_task(
     title: "Task description",
     priority: "P2",  // default, P1 if "urgent" mentioned
     pillar: "{from meeting}",
     people: ["{participants}"],
     source: "{meeting path}"
   )
   

   Copy

4. Mark as extracted by adding comment to meeting note:

   <!-- tasks-extracted: 2026-02-03T10:30:00Z -->
   

   Copy

Step 6: Auto-link People in Processed Notes

After finishing edits to each processed meeting note, run this once for every processed note:

node .scripts/auto-link-people.cjs "<note-file>"

Use node .scripts/auto-link-people.cjs --dry-run "<note-file>" to preview what would be linked without changing the file.

Step 7: Verify Entity Coverage

Run node .scripts/meeting-intel/verify-entities.cjs and show its one-line summary. If ENTITY_SUGGESTIONS_FILE contains suggested people, list them and ask: "Want me to create these pages? (creates via create_person; dismiss or never also fine)"

• Accepted: call create_person, set the suggestion to accepted, and set the contact state to created with its page path.
• Dismissed: set the suggestion to dismissed.
• Never: set the suggestion to suppressed.

Step 8: Summary Report

## Meeting Processing Complete ✅

**Synced meetings found:** X (last 7 days)
**Background sync status:** Running (last sync: 10 min ago)

### Updates Made

**Person pages:**
- Created: 3 new (Alice Chen, Bob Smith, Carol Wang)
- Updated: 5 existing

**Company pages:**
- Created: 1 new (Acme Corp)
- Updated: 2 existing

**Tasks extracted:** 7 items added to 03-Tasks/Tasks.md

### Recent Meetings

| Date | Meeting | Company | Participants |
|------|---------|---------|--------------|
| Feb 3 | Product Review | Acme | Alice, Bob |
| Feb 2 | Strategy Call | BigCo | Carol |

---
*Background sync runs every 30 min. Check status: `.scripts/meeting-intel/install-automation.sh --status`*

Error Handling

If no meetings found:

"No meetings synced in the last 7 days. Make sure:

1. Your Granola API key is connected (run /granola-setup if not)
2. Background sync is set up (run /process-meetings --setup)
3. Check logs: .scripts/logs/meeting-intel.stdout.log"

If background sync isn't running:

"Background sync appears to be stopped. To restart:

cd .scripts/meeting-intel && ./install-automation.sh
```"

Copy

Examples

/process-meetings

"Found 8 synced meetings. Updating 12 person pages, extracting 5 tasks..."

/process-meetings today

"Found 2 meetings from today. Processing..."

/process-meetings --setup

"Installing background automation..." [runs install script]

/process-meetings --people-only

"Updating person and company pages only (skipping task extraction)..."

Track Usage (Silent)

Update System/usage_log.md to mark meeting processing as used.

Analytics (Silent):

Call track_event with event_name meetings_processed and properties:

• meetings_count: number of meetings processed
• people_created: number of new person pages created
• todos_extracted: number of tasks extracted

This only fires if the user has opted into analytics. No action needed if it returns "analytics_disabled".