Know if your post will stop the scroll before you publish it.
Paste a LinkedIn draft, in French or English. Hookline reads it the way a stranger on a phone does and returns a /100 score explained pass by pass, a verdict on every sentence and five stronger hooks.
Harsh on purpose: the scale is anchored on what actually gets reach in 2026 (hook before "see more", paragraphs, no clichés, no bait). An average post scores like an average post.
02 · What you get back
41/100
Capped · a worn opener readers recognise as template talk
Capped · engagement bait, suppressed by the feed
- Hook · 30%
- 50
- Body · 25%
- 74
- Closing · 12%
- 40
Stronger hook
Posting every day cost me comments.
I measured it for 90 days.
Red = cut, amber = rework. Every verdict comes with its reason; every hook with the question it plants in the reader's head.
03 · How it judges
Cold reader first
A dedicated pass sees only your first 140 characters and the shape of the post, and says whether a stranger taps "see more".
Rules you can check
Walls of text, clichés, bait, hashtag stuffing and question openers are measured in code and cap the score. Each cap is named on the card.
Versions, not guesses
Edit, analyze again, compare two versions side by side and watch the score move. Six scored passes, weighted and explained.
04 · Pricing
- Free trial
- 3 analyses on sign-in with Google, no card
- Then
- Credit packs, one credit per analysis, no subscription
- Hooks
- One free regeneration per version, then one credit
- Failures
- Credits are reserved while it runs and only debited on success
- Your account
- No LinkedIn connection, no cookie, no automation: you paste text
05 · Built in public
Hookline is project 1 of a series by Sylvain Dendele, freelance backend developer: Django, DRF, Celery, usage-based billing, shipped in a week and documented day by day.