Market Signals: 30 Signals, 8 Categories
- The full intelligence terminal. Market data, protocol fundamentals, token unlocks, fundraising, governance, on-chain flows, sentiment, and AI research across every major asset and protocol.
- Pay in USDC for the exact call you need, or take one seat for $2,500 a year and stop counting.
Signal → trade and x402 volume measured —
What You Don't Know Is Already Costing You
Price is the last variable to reflect a change in market state. Volume, hourly activity, and weighted momentum move measurably earlier — but only if something is computing them continuously. Each definition below is a signal the agent recomputes every 30 minutes. Read: self-custody vs custodial AI trading agents →
Volume precedes price
VOLUME_ANOMALY = today's 24h volume ÷ trailing 7d average
Momentum is computed, not seen
TOKEN_PRICE_MOMENTUM = weighted 24h + 7d price change
A single feed can't validate itself
Cross-source divergence: Hyperliquid vs CoinGecko · 0.1% avg spread
Eight categories. Every asset class. Per call.
Everything is keyed to assets, tokens, and protocols. Every endpoint is callable on its own. Prices are per call in USDC via x402 — no account, no subscription.
What You Get Per Call
Representative calls from the catalog. Every response includes a freshness timestamp. Every predictive call includes a 90-day rolling hit-rate.
Forward call with horizon, drivers, and a published hit-rate. The hit-rate and sample size ship with every response — an agent can weight this signal, not just read it.
- ✓ Horizon and key drivers returned with every call
- ✓ 90-day rolling hit-rate included in every response
- ✓ Sample size returned — weight the signal, don't just read it
- ✗ Hit-rate is published, not adjusted retroactively
Labeled-wallet inflows and outflows. Accumulation and distribution from identifiable wallets, not aggregate flow. Sourced, not inferred.
- ✓ Per-wallet labeled inflows and outflows
- ✓ Accumulation vs distribution classified separately
- ✓ Identifiable wallets — sourced, not inferred
- ✗ Aggregate-only flow data will not show this
Vesting and emission schedule plus an impact forecast. Supply pressure quantified before the unlock date, not after the price has already moved.
- ✓ Full vesting and emission schedule per asset
- ✓ Impact forecast: expected supply pressure quantified
- ✓ Available before the unlock date
- ✗ Spot price alone does not show forward supply pressure
Any question, answered against live data, fully cited. Generated on demand — not a cached summary from last week.
- ✓ Live data at generation time — not a cached answer
- ✓ Fully cited — every claim is traceable
- ✓ Any question across any asset or protocol
- ✗ No static knowledge cut-off: yesterday's data answers today's question
Where the Intelligence Changes the Call
Four buyer types, four different decision problems. The catalog has the data each one needs, callable individually.
Position sizing that requires current data
The Problem
Predictive signals need a verifiable track record to size against. Smart-money flow needs to be sourced — not inferred from price — to confirm a thesis before a desk commits.
What the Catalog Returns
Every predictive call returns a 90-day rolling hit-rate and sample size. Smart-money flow is wallet-level, labeled, and sourced. Deep Research generates against live data on demand.
Fundamental research and supply-side risk
The Problem
P/S and P/E ratios require current revenue and fee data. Token unlock impact is a forward supply pressure calculation — not visible from price history alone.
What the Catalog Returns
Protocol Financials returns revenue, fees, TVL, and derived ratios. Token Unlocks returns the vesting schedule and an impact forecast. Governance tracks active proposals across protocols.
Data access without credential management
The Problem
An agent that needs data at execution time cannot pause for an OAuth flow or manage a rotating API key. The payment must be atomic and the response must be immediately actionable.
What the Catalog Returns
x402: the agent hits an endpoint, reads the price from the 402 response, pays USDC on Base, and gets data in the same round trip. No account, no API key, no session state. MCP server available.
Terminal-quality data without a terminal contract
The Problem
Institutional terminals cost $30,000–$150,000 a year. Most individual traders use a fraction of that data. A single metric should cost a cent; the full catalog should not require a sales call.
What the Catalog Returns
Asset Metrics at $0.01 per call. Alerts at $0.02 per fire. AI Copilot at $0.10 per query. The full catalog, unlimited, for $2,500 a year. No sales cycle.
The Hard Parts. Now Answerable.
Every analytical challenge has a measurable proxy. Each entry below maps a question that arises in practice to the catalog endpoint that answers it.
Catching a move before it's priced in
Volume expansion and momentum shifts precede price confirmation. By the time the candle is visible, the entry is late.
Catalog returns
VOLUME_ANOMALY returns the ratio of current 24h volume to the 7d average. PREDICTIVE_SIGNALS returns a forward call with horizon and drivers.
Sizing through a token unlock
A vesting event changes the float. The magnitude depends on who holds the vesting tokens, how long they've held them, and whether smart money is already distributing.
Catalog returns
TOKEN_UNLOCKS returns the schedule and an impact forecast. SMART_MONEY_FLOW shows whether labeled wallets are accumulating or distributing into the event.
Validating a thesis with current data
A research thesis built on data from last week may not survive this week's on-chain activity. Protocol revenue, developer commits, and governance votes all change.
Catalog returns
AI_COPILOT answers any question against live data, fully cited. DEEP_RESEARCH produces a long-form, sourced report on demand.
Tracking governance risk across a portfolio
A governance vote that passes can alter token economics, fee structures, or treasury allocations. Positions can be material before the vote outcome is reflected in price.
Catalog returns
GOVERNANCE returns active and passed proposals, outcomes, and voting data across protocols. WHALE_ALERTS flags large wallet activity near proposal deadlines.
Knowing when funded wallets are positioning
Price and volume signals are observable by anyone. Wallet-level flow from identifiable funded addresses is not — it is a leading indicator that runs ahead of price.
Catalog returns
SMART_MONEY_FLOW returns labeled-wallet inflows and outflows. SENTIMENT returns the narrative attention score and social momentum alongside the flow data.
7 Capabilities. 3 Are Ours Alone.
The three gaps below don't exist on any of the eight competing platforms we evaluated.
Prices are sourced from both Hyperliquid and CoinGecko simultaneously. Every signal is built on a price that has already passed a second opinion — 0.1% average divergence across all assets.
Every query returns a live Grade A quality score alongside the signal values. You know the confidence level of the data before you act on it, not after something goes wrong.
Signal weights update based on realized win-rate outcomes. Competitors ship static models and leave calibration to you. This system corrects itself with every completed cycle.
| Capability | This System | Best Competitor |
|---|---|---|
| HL-native OHLCVPrice candle data sourced directly from Hyperliquid | 30-min sync cycle | Aggregators — 24h+ feed lag, no dual-source |
| Pre-computed TA indicatorsTechnical indicators delivered ready-to-use | RSI · MACD · BB · ATR · EMA | Most providers — raw indicators, not pre-computed |
| L2 order book imbalance signalPre-computed directional pressure score | −1 to +1 pre-computed | Data providers — raw order book snapshots, no pre-computation |
| Autonomous HL DEX executionTrades placed directly on Hyperliquid | Native execution | Third-party integrations — no native HL execution |
| Dual-source cross-price validation Only usHL + CoinGecko agreement check on every price | HL + CG · 0.1% avg spread// second-opinion before every signal | None found |
| Live data quality scorecard Only usPer-query confidence score delivered with the signal | 97.9 / 100 · Grade A// included in every response | None found |
| Signal self-calibration loop Only usWin rate feedback updates signal weights automatically | Win rate feedback loop// model corrects with each cycle | None found |
A single metric costs a cent.
The whole platform costs $2,500 a year.
No five-figure contract. No sales cycle. No seat you barely use. Pay in USDC on Base via x402, or take one seat and stop counting.
No account. No API key. An agent hits an endpoint, reads the price from the 402 response, pays USDC on Base, and gets data in one round trip.
| Endpoint | Per call |
|---|---|
| Asset Metrics price, OHLCV, market cap, FDV, volume | $0.01 |
| Historical Timeseries any metric, full history | $0.01 |
| Derivatives funding, OI, basis, liquidations | $0.01 |
| Exchange & Market Data per-venue volume, spreads, depth + live order book | $0.01 |
| News & Trending asset-tagged feed with trending rank | $0.01 |
| Network Activity addresses, fees, developer commits | $0.02 |
| Stablecoins supply, mints/redemptions, per-chain flows | $0.02 |
| Staking & Yields staking rewards, DeFi pool APYs | $0.02 |
| Proposal Tracking live governance across protocols | $0.02 |
| Whale Alerts large transfers flagged in real time | $0.02 |
| Alerts any condition — webhook / Telegram / push | $0.02 / fire |
| Protocol Financials revenue, fees, TVL, P/S, P/E | $0.05 |
| Treasury treasury composition and runway | $0.05 |
| Screener filter and rank 500+ assets | $0.05 |
| Watchlists + AI Digest curated lists, daily/weekly AI summaries | $0.05 |
| Sentiment & Mindshare social sentiment, narrative score | $0.05 |
| Market Signals momentum, volume anomaly, technicals | $0.10 |
| Smart-Money Flow labeled-wallet inflows / outflows | $0.10 |
| AI Copilot cited answer on live data | $0.10 |
| Fundraising rounds, investors, valuations | $0.10 |
| Predictive Signals forward call + published 90d hit-rate | $0.25 |
| Token Unlocks vesting schedule + impact forecast | $0.25 |
| Deep Research async, long-form, fully cited report | $2.00 |
| Asset Diligence standardized report — tech, tokenomics, risks | $2.00 |
| Sector & Thematic Research narrative and sector-level reports | $2.00 |
| Cohort Intelligence snapshot → competitive intel, 5 tiers | $0.50–$25 |
- Every metric, signal, report, and alert
- No per-call metering — no counting
- AI Copilot and Deep Research on demand
- Predictive Signals with 90d rolling hit-rate
- Token Unlock impact forecasts
- Smart-Money Flow labeled wallet tracking
- Pay once in USDC — no sales call
Pay for the one thing you need, or take a seat and stop counting.
One round trip. No account.
x402 is a standard HTTP extension. Any HTTP client pays and receives data in a single request. An MCP server is available for agent frameworks that prefer tool-call integration. Full integration guide: api.fexr.io/x402
GET api.fexr.io/v1/data/market-signals?assets=BTC
← 402 Payment Required · x402 challenge
"maxAmountRequired": "100000" // $0.10 USDC
"payTo": "0x…" · "chainId": 8453
→ pay USDC on Base, retry with X-PAYMENT header
GET api.fexr.io/v1/data/market-signals?assets=BTC
X-PAYMENT: base64(receipt)
← 200 OK — provenance-wrapped data
{
"value": { "momentum": 0.62, "rsi_14": 50.1, … },
"fresh": true, "age_seconds": 232,
"source_status": "ok"
}
Also available as an MCP server
28 typed tools, generated from the live price registry — prices embedded in every tool description so agents can budget before they call. Server: api.fexr.io/v1/mcp
Everything the terminals give you —
fresher, cheaper, and callable by the agent.
Pull exactly what you need for a cent, or take the whole platform for $2,500 a year.