# protocol exploit — X 热门讨论 (2026-09-22 01:06 UTC)
## @DefiLlama (DefiLlama.com) · 09-21 22:05 · ♥21 ↻3 💬7 This prompt will change how you do crypto research. Paste it into DefiLlama's LlamaAI to find trending tokens, assess their fundamentals, check for red flags, and look up what people are saying about them.
Find tokens trending in the last 24-48 hours using at least two independent signals: news/narrative mention spikes against baseline, price/volume/mcap movers, and X mention volume with sentiment. Require corroboration across at least two signals per token, and note which signals flagged each one. Return 8-12 candidates, spread across categories rather than one narrative dominating the list. Separate large-cap tokens (top 100 by market cap) from smaller or emerging ones, since "trending" means something different for each.
Before pulling any fundamental figures, resolve each candidate's identity precisely: confirm whether it is a chain-native token, a protocol fee/governance token, or a token with no underlying protocol at all. If a symbol has more than one live contract, identify which one the trending signal actually refers to before analyzing anything. If the correct contract cannot be established with confidence, do not publish fundamentals for it: drop it from the ranked table and list it separately as identity-unresolved, with what would resolve it.
For each resolved candidate, state in one sentence what the project does, its category, and its chain(s).
Assess fundamentals only where they exist: - Price, market cap, fully diluted valuation, and the gap between them - For protocol or chain tokens: TVL, fees, revenue, and price-to-sales / price-to-fees ratios against category peers - 7-day, 30-day, and 90-day price and volume change, checked against TVL/fee change over the same windows, to separate price-driven moves from usage-driven ones - Any unlock cliff in the next 30-90 days and its size relative to circulating supply - Universal Token Rating, if the token has one
For tokens with no protocol behind them, skip the fields that do not apply and say so rather than forcing a number in.
Assess social activity separately from fundamentals: - Mention volume trend and sentiment direction on X over the past 24-72 hours - Whether a small number of accounts are driving most of the volume, or it is broad-based - Whether social attention is rising faster than, in line with, or behind the fundamental data above
Check for red flags before writing a verdict: concentrated holder base, a recent large unlock already sold into the market, known contract or audit issues, and any active exploit or governance dispute. When two sources disagree materially on a red-flag figure, say which one you trust more and why, or mark the figure unresolved-conflicting; never present two contradictory numbers as if either could be picked at random.
Write one row per token: name, category, what triggered the trending flag, the fundamental read, the social read, red flags found, and a verdict. Use one of four verdicts, each tied to a rule: fundamentally-backed (usage or fees moving with price), narrative-led (social and price moving, fundamentals flat), unlock-risk (fundamentals fine, but a near-term unlock threatens the float), or unverifiable (identity or data conflict prevents a real read, distinct from identity-unresolved candidates already excluded above). Rank by verdict strength, not by how hard a token is trending. State the source next to every figure.
If you chart price changes across the candidate set and one token's move is an order of magnitude larger than the rest, exclude it from that shared chart (or give it a log-scale or separate chart) so the other bars stay readable; keep the number in the table regardless.
Close by naming the two or three tokens most worth a deeper look, and for each, the specific data point that would confirm or kill the thesis. https://x.com/DefiLlama/status/2102157171530121217