Painfinder mines Reddit for your niche and returns a ranked list of what users actually complain about — with verbatim quotes, source links, and a product opportunity for each theme. 5 minutes, one command.
A markdown report with 5–10 ranked pain themes for any niche. Each theme includes:
✅ The pain in one sentence
✅ Who is affected
✅ Verbatim evidence — real quotes from real threads
✅ Source links — click straight through
✅ Frequency (frequent vs scattered)
✅ An implied product opportunity
FREQUENT Dental PMS software is buggy and fragmented
OPPORTUNITY Build a modern, cloud-based dental PMS that's actually usable.
FREQUENT No automation for recalls / patient reactivation
OPPORTUNITY Automated recall & reactivation workflows for front desks.
SCATTERED Imaging software switchover pain
OPPORTUNITY Imaging software with drop-in integration, zero retraining.
git clone https://github.com/kragent66-glitch/painfinder
cd painfinder
pip install -r requirements.txt
echo "DEEPSEEK_API_KEY=sk-..." > .env # get one at platform.deepseek.com
python3 painfinder.py "dental practice software"
# → reports/dental-practice-software-2026-08-11.md
Requires Python 3.10+, rdt-cli (one-time rdt login), and a DeepSeek API key (pay-as-you-go, fractions of a cent per niche). Full docs in the README.
12 pain-signal searches per niche (annoying, hate, expensive, sucks, alternative…) + optional subreddit front pages.
Drops removed posts, self-promo, non-English, memes. Keeps only relevant threads.
One LLM pass groups complaints into 5–10 pain themes with verbatim evidence.
Ranked markdown report with sources + JSON dump for auditing.
I'll run painfinder on your niche and send back the full report.
You pay only the API-token cost — no markup, no service fee.