How to use this page
The /price page is a scanner with more controls than a typical screener — Trend intervals, story counts, thesis counts, fragility scores, sector filters, and numeric screeners — and the interactions between those controls are where the real power lives. Each use case below is self-contained: a specific goal, the exact steps, and an explanation of why the approach works.
Real-time narrative discovery
Goal: Find tokens generating on-chain stories right now.
Set the Trend picker to 1H and sort Stories descending. The tokens at the top are where the agent created the most stories in the last sixty minutes — active on-chain events like whale movements, cluster formations, surge detections, and liquidity shifts happening now. This is like switching from a daily weather forecast to a live storm-tracking radar. The 1H Stories count reflects agent output refreshed on a 10-minute ClickHouse cache, giving you a near-real-time ranking of narrative attention across 7,000+ tokens.
Daily conviction scan
Goal: Find tokens where the AI formed a directional view today.
Set Trend to 24H and sort Theses descending. The tokens at the top have active theses — structured LONG, SHORT, or AVOID briefs — generated in the last 24 hours. Theses are the agent's highest-conviction output, requiring multiple independent on-chain signals to converge. This is the difference between reading a stream of police reports (stories) and reading the detective's case summary (thesis).
Narrative + conviction filter
Goal: Find tokens with BOTH story activity AND a thesis — the highest-signal combination.
Set Trend to 24H, sort Theses descending, and scan for rows where both Stories and Theses are non-zero. These tokens have the full AI coverage stack: raw observations are ongoing AND evidence has converged into a directional view. This is the screener's most powerful compound signal — the equivalent of finding both smoke and the fire alarm confirming the smoke is worth your attention.
Sector rotation scan
Goal: Which sector is generating the most narrative activity today?
Use the category filter to select a sector (DeFi, L2, Meme, etc.), set Trend to 24H, sort Stories descending, and compare top-end story counts across sectors. On-chain story activity often precedes price movement because the events the agent detects — whale accumulations, cluster formations, liquidity shifts — are the mechanical causes of price changes. This is like checking which neighborhoods have the most lights switching on before the crowd forms.
Hidden gem discovery
Goal: Find small-cap tokens the agent is already watching.
Open the Screener panel and set a maximum MCap (e.g. $50M for micro-cap). Set Trend to 24H and sort Stories descending. The agent covers 7,000+ tokens with no size bias — a micro-cap experiencing genuine whale accumulation shows up just as clearly as Ethereum. This is sweeping a metal detector across the whole beach instead of standing where everyone else is already digging.
Fragility + narrative cross-check
Goal: Spot tokens with activity that may be structurally weak.
Sort Stories descending and scan the Fragility column on the top results. High fragility (60+) combined with high story count is a warning sign — the on-chain activity might be real, but the token's price, MCap, and liquidity data cannot be fully trusted. Think of it as a stress test on the dashboard itself: a bridge can look busy but be swaying under load. Fragility is the sway measurement.