Today's Hunter verdict

A baseline snapshot is captured at 00:00 WIB each day. Every subsequent tick we diff the current state against that baseline and surface what actually changed: regime flips, new leading narrative, smart money rotations, new REAL CT terms, concentration alarm states. Deterministic diff — no LLM guessing.

Read order. The top line is the one-sentence editorial summary. Below it are the individual changes ordered by severity: major first, then notable, then minor. If the day is genuinely quiet, we say so honestly instead of padding the feed.

Use. Open this first in the morning — it tells you what to re-check vs. yesterday's plan. If it says "quiet, no material change," trust your existing positioning and don't trade the noise.

Synthesized whale flow from QuickNode. Every 3 minutes we read Uniswap V3 Swap events on the three deepest Ethereum pools (WETH/USDC, WETH/USDT, WBTC/WETH), keep swaps ≥ $25K, and aggregate into the metrics shown. The hero gauge is the single question a trader asks first: are whales net buying or selling right now?

Net flow = Σ buy-side USD − Σ sell-side USD over the last ~50 min. Positive + aqua = accumulation; negative + rose = distribution. Buy pressure = buy share of total USD volume (50% = balanced, 70%+ = one-sided regime).

Per-token rows surface which specific asset the flow is in. Tempo is the 10-min swap count over the last hour — rising bars = activity is accelerating, flat/falling = cooling off. Clustered same-direction flow (3+ WHALEs one-side in 30 min) is the actionable regime signal.

A narrative/rotation forecast synthesized by our AI stack. Reads the current Hunter state and makes falsifiable calls on (a) which narrative leads today + this week, (b) which major coin (BTC/ETH/SOL) outperforms, (c) the single top gainer, (d) whether smart money stays bid on its current pick, (e) which rotation play hits +5% outperformance over BTC in 7d, and (f) a bold call for the week.

Accountability. Every field is graded automatically at T+1 or T+7 by the resolver — accuracy per field is surfaced below. This is a scoreboard, not a demo: if a call misses, it shows up red next time.

Don't expect magic. Single-day directional calls on crypto are noisy. The useful signal is (i) the model's own calibration over time and (ii) cross-checking the model's pick against your own read of the underlying panels.

A single AI synthesis across every Hunter panel. Reads morning_verdict, smart_money_now, ct_emergence, narrative_entries, rotation_plays, chase_risk, and concentration_alarm — then returns a tight verdict, a few plays to consider, a few to avoid, and the biggest risk to the thesis.

How to read. Regime + confidence pill sets posture. Thesis paragraph tells you what to do. Play cards link to tokens that appear in at least one panel — no invented tickers. Avoid cards surface the chase-risk + HYPE-flagged terms.

How to use. Start every session here. If the thesis disagrees with your own read, check the underlying panels to find where. Regenerates 4×/day (00/06/12/18 UTC) after Grok + Nansen refresh.

The best single entry candidate on the board today. Scores every narrative for "mindshare is leading price" (the textbook early-entry pattern), picks the winner, then picks the dominant ticker inside it. Answers Q5 (how early is early enough) and Q11 (mindshare up 40% but price only up 5%).

How to read. SEEDLING = fresh, nobody in yet. GROWING = momentum building, the zone. PEAK = crowded trade, late. DECAYING = attention leaving. DORMANT = no signal. Top pick is the ticker with dominant mindshare within the narrative. Phase summary tells you how crowded the trade is.

How to use. Enter on SEEDLING or early GROWING. Skip PEAK (you're late). Don't chase DECAYING. Pair with Rotation Plays to pick the right ticker once you like the narrative.

The whole narrative landscape by lifecycle stage. Early Rotation picks the single best entry; Phase Map zooms out — are we in broad accumulation, late-cycle distribution, or mixed? The macro read for "should I be aggressive or selective today?"

How to read. Each zone groups narratives by stage. Seedling = fresh, no crowd. Growing = momentum building, the hunt zone. Peak = crowded. Decaying = attention leaving. Dormant narratives are listed compactly so you know what's NOT moving.

How to use. Growing count high + Decaying low = go hunt. Peak count high + Decaying rising = be selective, late-cycle. Dormant dominant = sit out, nothing to play.

Which narratives are hottest right now? A 0–100 composite score combining mindshare (25%), 7d price change (30%), edge scores (25%), and token breadth (20%), multiplied by lifecycle stage. ACCELERATION narratives get a 1.3x boost; DORMANT get 0.1x.

How to read. Heat ≥60 = strong conviction zone. 30–60 = watch closely. <30 = cooling or dormant. Correlation penalty deducts points if the narrative moves identically to others (low uniqueness).

Which narratives move together? Cross-correlates 7d token returns across narrative pairs. If you're long AI + AI Agents, you might be holding the same risk twice. Diverging pairs flag potential rotation opportunities.

How to read. STRONG correlation (r≥0.7) = highly similar price moves, check for token overlap. MODERATE (r≥0.4) = some co-movement. Diverging = moving in opposite directions.

The shortest-horizon signal we have. Every 15 min we pull Birdeye's trending lists for Solana + Ethereum + Base, then enrich each with 30-minute / 1-hour / 4-hour price deltas and unique-wallet counts. A ticker lands here only if it's both trending AND has fresh buyers arriving in the last hour — that's the pre-pump window.

How to read. Score blends 30m price, 1h wallet growth, liquidity, and trending rank. Top rows are the tokens with the most momentum right now, not the most momentum 24h ago (which is what chase_risk, matrix, and most of the rest of this product shows).

How to use. Paired with chase_risk and rotation_plays — if a ticker shows CHASE_OK from chase_risk AND appears here, that's two independent confirmations. Alone, treat as watchlist ideas, not signals — sub-hour movers reverse fast.

A unified timeline of early attention and accumulation. Merges Emerging Tickers, New on CT, and Pre-Pump LP into a single feed. Filter by source to isolate specific signal types.

How to read. Items are tagged by source: ⚡ LP (Smart money liquidity add), 🧑‍💻 CT (Mindshare surge), 📊 MINT (Fresh contract launch). Sort by time or relevance.

The full matrix as a scannable table. Sort any column, filter by narrative, set a minimum composite score, or search by ticker. Every row is the same data the other Hunter panels see — just laid out so you can answer "show me X tokens sorted by Y" in one glance.

How to read. CS = composite score (0-10). MS = mindshare score (0-1). 24h / 7d = price change. MCAP = market cap. Click any column header to toggle sort direction. Narrative dropdown filters to one narrative at a time.

How to use. Research mode — not a trading signal. Pair with the Confluence Score drawer below for any ticker that catches your eye.

A unified 0-10 Confluence Score per token. Merges Confirmation Board, Divergence Watch, Whale Cohort, and Cut Watch into a single actionable card.

How to read. Shows 5 check-chips: Mindshare, Attention Gap, Smart Money Flow, DEX Liquidity, and Momentum. Overall verdict determines if a token is a GO, CAUTION, or EXIT.

One curated sentence from the Nansen SM tape, every 15 minutes. Picks the highest-surprise signal across the feed — rotation, clustering, conviction, sign-flip, or net-bias — and surfaces only that. Raw tape and per-token holdings exist at the API level for audit but aren't re-listed here. Our edge is synthesis, not re-printing Nansen's index.

How to read. The headline is the takeaway. The evidence line is the hard data behind it (wallet counts, position sizes, net bias). If it says "rotation" you've got capital moving between tokens. "Conviction" means sizes are scaling up on one side. "Sign-flip" = the cohort just changed direction.

How to use. A pulse check. "What is the room doing that I'm not doing yet?" Cross-reference Early Rotation and Divergence Watch before sizing a trade.

"The token pumped 30% in 4 hours — chase or wait?" The playbook default is "wait for pullback". This panel takes today's pumpers (≥15% 24h) and grades each on whether the move earned the chase. Answers Q13.

How to read. CHASE OK = mindshare caught up to price AND 7d still under 50%. Reasonable entry on a pullback. WAIT = price ran ahead of attention — unsustainable, wait for retrace. TOO LATE = ≥80% on the week or ≥60% in 24h, you're buying the tail.

How to use. CHASE OK → plan a scaled entry on the next retrace. WAIT → do nothing, revisit later. TOO LATE → stop reading and find a laggard in Rotation Plays instead.

Per-altcoin funding rates across exchanges. Shows the top 30 tokens with the most extreme funding rates (positive = longs paying, negative = shorts paying). Aggregated from Coinglass across 20+ exchanges, stablecoin margin only.

How to read. EXTREME LONG = avg funding >1% — overcrowded long, expect a flush. LONG BIAS = moderate long positioning. SHORT BIAS = shorts paying — potential short squeeze. EXTREME SHORT = deeply negative — strong squeeze candidate.

How to use. Cross-reference with Narrative Phase Map — a narrative in PEAK with extreme long funding is a distribution setup. A GROWING narrative with negative funding is the cleanest entry.

Contract-safety pass on every active alpha candidate. Runs GoPlus Token Security across the tickers Early Rotation / Rotation Plays / Basket Builder / Confirmation Board / CT Emergence / Chase Risk are surfacing right now. Hard-filter for honeypots, pausable transfers, blacklist functions, abusive sell taxes, mint authorities, and rug-history creators. Answers Q22.

How to read. Each card shows ticker, chain, and a verdict. SAFE = no rug primitives detected. CAUTION = soft warnings only (mintable owner, proxy contract, buy tax, unverified). AVOID = hard-fail detected (honeypot / sell tax ≥10% / pausable / blacklist / hidden owner / slippage-modifiable / selfdestruct / creator rug history). UNKNOWN = GoPlus has no data (token too fresh). Issue chips at the bottom show the specific findings.

How to use. AVOID is a hard block — do not enter that position regardless of signal strength on other panels. CAUTION = read the specific warnings and size accordingly (mintable alone is common for legit majors; pausable is usually a red flag). UNRESOLVED list at bottom shows tokens we can't auto-screen (no contract in SM holdings) — check those manually on goplus.io.

Are you about to pile into one narrative without noticing? Reads every active pick across Early Rotation, Rotation Plays, Confirmation Board (CONFIRMED/MIXED), CT Emergence (REAL/MIXED), and Chase Risk (CHASE_OK), groups them by narrative, and tells you if the hunt is diversified or all in one theme. Answers Q23.

How to read. DIVERSE = ≥3 narratives, top <50%. Healthy. BALANCED = 2-3 narratives, top <65%. Acceptable. CONCENTRATED = top narrative ≥65%. Could be a conviction bet, could be accidental. OVER_INDEXED = top narrative ≥85% or only 1 narrative. Rotate capital out. Bar chart shows pick count per narrative, dominant narrative highlighted.

How to use. If the verdict is OVER_INDEXED or CONCENTRATED and you hadn't planned it, pull one position out of the dominant theme and redeploy into a laggard from a different narrative (use Rotation Plays to find one). If it's your conviction play, size accordingly but know you're making a single bet.

"Narrative is real but every token already ran 3x — where do I enter?" For each GROWING or PEAK narrative, the leader (where the herd already is) plus laggards in the same narrative that haven't moved yet but still carry attention. Second-wave entries. Answers Q10.

How to read. Leader chip = ticker with dominant mindshare inside the narrative; usually already ran. Laggard chips = same-narrative tickers with 7d ≤ +20% AND mindshare floor ≥ 0.003. Each chip shows $TICKER, 7d %, and mindshare. GROWING narratives are listed first (more upside), then PEAK.

How to use. Skip the leader (priced in). Buy laggards, sized by composite score order (first chip = best). Confirm the laggard against Chase Risk (shouldn't be TOO_LATE) and Cut Watch (shouldn't already be stalling).

Every tracked narrative plotted on one 2×2. X-axis = 7d price change (where it's been). Y-axis = 24h price change (what it's doing right now). The four quadrants tell you whether each narrative is accelerating, flipping bullish, cooling, or bleeding.

How to read. Q1 Accelerating (↑7d ↑24h) = both trends up, momentum confirmed. Q2 Flipping bullish (↓7d ↑24h) = bouncing off weakness, potential reversal. Q4 Cooling (↑7d ↓24h) = was hot, losing steam. Q3 Bleeding (↓7d ↓24h) = both directions down. Dot size = narrative mindshare.

How to use. Q1 = follow momentum. Q2 = early reversal, higher risk but more upside. Q4 = start taking profits. Q3 = avoid. Cross-ref with Narrative Phase Map to see lifecycle stage alongside price direction.

Does the Edge Score actually predict anything? Two horizons of forward-looking performance for the top-5 Edge tickers: 24h and 7d. If yesterday's top composite scores went up today, the model has signal. If they didn't, we're guessing and you should weight your own judgment higher.

How to read. Each card shows the pick date, then the 5 top-composite tickers from that snapshot with their forward return measured against the current matrix. Aqua = positive return (hit). Rose = negative (miss). Summary shows hit rate, median return, and the best/worst pick. "Accumulating" means we don't have enough snapshot history yet to compute that horizon.

How to use. Sanity check before trusting today's Spotlight. If last 24h hit-rate is 4/5 with +8% median, the model is on. If it's 1/5 with -10% median, fade your conviction — favor your own reads over the scores today. 7d horizon unlocks once we accumulate a week of snapshots.

Methodology. "Forward return" = the current matrix's price_change_24h_pct or price_change_7d_pct for each pick. Because the snapshot is exactly ~N hours old, that equals the return from pick-time to now. Honest, no external price API needed.

Every pick Hunter makes is logged and graded. This is the permanent record of rotation laggards, CT emergence calls, and smart money signals — each measured against their forward return from pick price to current price. The system cannot cherry-pick winners or hide losses. Sprint 4 accountability layer.

How to read. HIT = positive forward return from pick date. MISS = negative return. Hit Rate = % of picks that went up. Avg Return = mean forward return across all picks.

How do Hunter's picks perform under different market regimes? This panel breaks down forward returns by narrative and signal source, conditioned on the active regime when the pick was made. It answers: "when the regime is HUNT, which narratives do best?" Sprint 4 whitespace gap #4.

How to read. Each row shows a narrative or signal module with its hit rate and average return. As more data accumulates across HUNT/CAUTION/SIT_OUT regimes, regime-specific breakdowns will appear.

Pick the narrative right, still pick the wrong token? Per hot narrative, shows two ways to express the thesis: a single conviction pick (the leader) or an equal-weight basket (leader + laggards). The card tells you which one the data prefers. Answers Q9.

How to read. PREFER CONVICTION = leader has ≥3× the mindshare of the next name — concentration matches dominance, basket would dilute. PREFER BASKET = leader already +25% 7d + laggards still flat — leader priced in, basket catches second-wave rotation. EITHER WORKS = no extreme signal, pick on sizing preference. Basket stats show avg 7d return + average mindshare across the N tokens at equal weight.

How to use. If PREFER CONVICTION, size one position on the leader. If PREFER BASKET, split capital equal-weight across all N tokens — one of them wins, and idiosyncratic "wrong token" risk collapses. Combine with Concentration Alarm: if you're already OVER_INDEXED, basket doesn't help — rotate to a different narrative.

Universe-level scoreboard. Tokens tracked in the matrix, stage counts, emerging count, bridge flow totals. Not a trading signal — a "how broad is the opportunity set today" read.

How to read. Totals = raw coverage numbers. Growing/Peak counts = where attention is concentrated today. Bridges = macro chain rotation going on in the background.

How to use. A gut-check before diving deeper. Very low Growing + high Dormant = quiet market, don't force trades. High Growing + multiple narratives = rotation-rich day, pick carefully.