How It Works
There’s no wall to climb over and no willpower to run out of. There’s a scoreboard, and there’s one person who can see it. That’s the whole app.
You name a Clamp — the one person you’d least like to disappoint. The app watches for flagged activity. Your Clamp finds out. You can delete an app in a weak moment; you cannot un-tell someone who already knows.
What counts as an event
The app sorts activity into a category right on your phone, and every flagged event gets a severity: Tier I, Tier II, or Tier III. Tier I and II wait in line for the weekly bulletin. Tier III does not wait — your Clamp gets an email the moment it lands. Some things can’t sit until Sunday.
The Weekly Risk Classification
Every week your Clamp gets one number: where you landed on the risk scale. Five rungs, lowest to highest. Zero is a rung too — a clean week is Diamond Hands, which is also the only rung we’ll quietly admire and never tell you about.
Thresholds shown are current targets and will be tuned as the app meets real-world data. Calibration is the responsibility of Ben Overbich, Tier Threshold Calibration Lead (ext. 4410). He has assured us the numbers are correct. We have elected to believe him.
The Rank Ladder
Stay clean and you build a streak. Your streak earns you a rank. The ranks climb. The names do not get more flattering.
- 21+ days Highly Regarded the top rank — yes, really. Our highest classification, despite sounding like an insult.
- 8–20 days Ape
- 4–7 days Smooth Brain
- 0–3 days Degenerate where everyone starts, and where everyone returns.
Any incident resets your streak to zero. No partial credit. You start the climb again from Degenerate. We attempted to build in a grace day. We were unable.
What your Clamp actually sees
Your accountability partner sees how many events occurred in each category, never which sites or what was on them.
ENC monitors content in system browsers (Chrome, Safari, Firefox, Edge). Content viewed in in-app browsers — Reddit, Instagram, X, Facebook, Pinterest, TikTok, and similar — is not monitored in this version. This limitation is disclosed proactively; an expanded coverage scope is on the roadmap.
Classification uses OpenAI Moderation (text) and Google Cloud Vision SafeSearch (images), configured for zero data retention. These services see flagged content briefly for classification; ENC’s servers receive only the resulting category labels. Neither receives your identity or any link to your account.