Another day, another SaaS platform holding my data hostage behind a pretty UI and a "Download" button that only exports the last 30 days. Fellow, I'm looking at you. Your action items feature is genuinely useful for tracking who owes me what after a meeting, but trying to get a historical view for any kind of FinOps or process audit is an exercise in frustration. Want to see if your team's action completion rate improved last quarter? Good luck manually paginating and copying.
So I did what any engineer tired of marketing claims about "open platforms" would do: I poked at the network tab. Turns that while Fellow doesn't provide a bulk export, their API is at least semi-competent. I've written a script to crawl all your action items—yours, those assigned to you, all of them—and dump them into a sane, queryable CSV file. It's not pretty, but it works.
You'll need a Fellow API token, which you can get from your settings page. The script handles pagination and flattens the nested JSON into something you can throw into a spreadsheet. It also pulls in some linked meeting data, because an action item without its meeting context is just a sad, orphaned todo.
```python
import requests
import csv
import time
import sys
# Configure
API_TOKEN = 'YOUR_TOKEN_HERE' # From Fellow Settings -> API Integration
BASE_URL = 'https://api.fellow.app/v1'
HEADERS = {
'Authorization': f'Bearer {API_TOKEN}',
'Content-Type': 'application/json',
}
def get_all_action_items():
"""Paginate through all action items."""
items = []
url = f'{BASE_URL}/me/actionitems'
page = 1
while url:
print(f"Fetching page {page}...", file=sys.stderr)
resp = requests.get(url, headers=HEADERS)
if resp.status_code != 200:
print(f"Error: {resp.status_code} - {resp.text}", file=sys.stderr)
break
data = resp.json()
items.extend(data.get('actionItems', []))
url = data.get('pagination', {}).get('next')
page += 1
time.sleep(0.1) # Be polite to their API
return items
def flatten_action_item(item):
"""Extract the fields we actually care about."""
meeting = item.get('meeting', {})
return {
'id': item.get('id'),
'created_at': item.get('createdAt'),
'updated_at': item.get('updatedAt'),
'body': item.get('body'),
'status': item.get('status'),
'due_date': item.get('dueDate'),
'completion_time': item.get('completionTime'),
'owner_id': item.get('owner', {}).get('id'),
'owner_email': item.get('owner', {}).get('email'),
'assignee_id': item.get('assignee', {}).get('id'),
'assignee_email': item.get('assignee', {}).get('email'),
'meeting_id': meeting.get('id'),
'meeting_title': meeting.get('title'),
'meeting_start_time': meeting.get('startTime'),
}
def main():
items = get_all_action_items()
if not items:
print("No action items found.", file=sys.stderr)
return
fieldnames = [
'id', 'created_at', 'updated_at', 'body', 'status', 'due_date',
'completion_time', 'owner_id', 'owner_email', 'assignee_id',
'assignee_email', 'meeting_id', 'meeting_title', 'meeting_start_time'
]
with open('fellow_action_items.csv', 'w', newline='', encoding='utf-8') as f:
writer = csv.DictWriter(f, fieldnames=fieldnames)
writer.writeheader()
for item in items:
writer.writerow(flatten_action_item(item))
print(f"Exported {len(items)} action items to fellow_action_items.csv", file=sys.stderr)
if __name__ == '__main__':
main()
```
Run it, get your CSV, and analyze to your heart's content. I've used this to identify recurring bottlenecks (looking at you, "follow up with legal") and to shame my own team with completion statistics. It's baffling that this isn't a built-in feature for a tool that charges per seat. The script is a stopgap, but it's better than the alternative of being locked in.
A few caveats: The API rate limits aren't documented, so the script sleeps between calls. It only fetches action items *you* have access to. And as with any unofficial script, it'll break when Fellow decides to "improve" their API without notice. But until then, enjoy actually owning your data.
-- Cam
Trust but verify.