Route an x402 buyer or autonomous agent from a plain-language crypto intelligence goal to the most useful TradeOS service. Returns recommended free and paid SKUs, reasons, exact request examples, pricing, output expectations, buyer-safe trust summaries, and a feedback request block. This is a paid x402 discovery surface for marketplace indexing and conversion; deeper signal evidence, risk, dataset, and order-flow intelligence stays in the dedicated SKUs.

One TradeOS entry point for crypto intelligence when a buyer does not want to choose a specialized endpoint first. The router returns read-only JSON from token risk, discovery, market pulse, current signal candidates, signal evidence, forecast context, order-flow stress, or dataset scoping. Built for agents, apps, and dashboards that need plain reasons, freshness, filters, and safety boundaries without trade execution.

Paid cycle radar for token watchlists and research agents. Returns a plain consumer read such as Review Buy, Review Sell, Possible Bottom, Possible Top, Mixed / Unclear, or Avoid, using a 30D Research Read, plus raw bottom/top fields, risk context, projected-target context, explicit scam/risk blocking, warnings, and freshness so downstream agents can route tokens to review queues without execution or personalized financial advice. Projected targets may carry separate 30D or 90D target context.

Help a buyer choose the right TradeOS dataset package before a private delivery. Returns recommended replay, forecast-checking, decision, risk-event, and evaluation packages with fit, field hints, freshness, and access path.

Create a dataset package preview before private delivery. Returns the time window, covered symbols, expected fields, caveats, access-review fields, and delivery readiness.

Explain how a forecast is tracking against its target zone over 1d, 3d, and 7d review windows. Returns target path, follow-up status, confidence, market caveats, and invalidation evidence without giving a trade order.

Crypto Signal Snapshot returns up to 25 current signal candidates from TradeOS for agents, dashboards, and screeners. It includes reasons, filters, freshness, and safety boundaries so a buyer can refresh a watchlist without cleaning raw feeds or receiving trade instructions. Read-only research output only; no execution, custody, copy trading, or personalized financial advice.

A one-day signal history window for dashboard backfills, recent behavior audits, and agents that need context beyond the latest snapshot. It returns up to 200 sanitized historical rows with filters, timestamps, scores, decision states, and freshness metadata, so consumers can replay what changed without buying or reselling a raw feed. Read-only history, not execution.

A seven-day signal history window for research audits, dashboards, and model evaluation. It returns up to 500 sanitized historical rows with filters, timestamps, scores, decision states, and freshness metadata. Consumers get a bounded weekly review window for agents and analysts; it is not copy trading, raw-feed resale, or execution.

OpenAI GPT OSS 120B raw inference hosted on Agentic Source Lab Spark-2 and exposed through the TradeOS OpenAI-compatible, AntSeed, x402, and Virtuals ACP surfaces. This SKU is optimized for agent builders that need a recognizable open-weight reasoning/coding model without managing their own GPU host. It is model output only, not a TradeOS evidence, signal, risk, market, order-flow, or dataset claim.

Team/API package for agents that need one TradeOS context call across market reads, token risk/discovery, signal quality/evidence, forecast context, order-flow stress, Market Pulse Intelligence v2, VPIN flow-risk v2 incident evidence, dataset coverage, and optional portfolio context.

Paid real-time market and sector context for agents. Returns global read, strongest tokens, weakest tokens, why-it-says-that inputs, coverage, optional portfolio lens, risk controls, Market Pulse Intelligence v2 shadow fields, VPIN flow-risk v2 incident evidence, and structured market-state fields.

Run discovery plus token-risk, market-read, and order-flow stress checks in one paid response. Designed for scanner agents that need candidates already filtered for obvious scam evidence, liquidity stress, unsafe market conditions, and recent VPIN shock follow-through. This is a plain risk gate for discovery workflows, not a trade instruction.

Plain-English reasoning for agents and dashboards that need to understand why a signal passed, paused, or was blocked. The response explains reliability drivers, disagreement, forecast alignment, risk flags, and safe next checks. Designed for validation workflows that need reasoning, not a bare score or trade instruction.

A review-ready evidence packet for bots, dashboards, and analysts that need to understand why a crypto signal deserves attention. It joins the current signal state with forecast context, market checks, order-flow stress, VPIN shock incident follow-through, freshness, disagreement drivers, and risk flags. Built for validation queues that need evidence, not another unexplained long/short call.

Validate an existing TradeOS signal using recent outcome evidence. Returns reliability, target progress, sample context, readiness labels, and caveats for dashboards, confidence filters, and strategy validation.

Find Base, Solana, and onchain candidates for research watchlists. Returns ranked candidates with risk, a plain at-a-glance read, explicit scam/risk gate labels, target window, reachability, confidence, exit-ease, watch-for context, and missing-data flags. Long Term means a 90D crypto thesis; Short Term means a 30D crypto setup. No buy/sell calls.

Screen a token contract or symbol before it enters a bot, app, or research watchlist. Returns a read-only risk verdict with explicit scam callouts, smart-contract, liquidity, holder-concentration, token-structure, value-capture, confidence, missing-data, evidence graph, and next-check fields. Positioned as a token risk pay-per-call API and agent token risk gate, not a trading endpoint.

Discounted Venice-backed raw inference route for `claude-opus-4-5` through the TradeOS AntSeed/OpenAI-compatible surface. This is model output only, not a TradeOS evidence, signal, risk, market, order-flow, or dataset claim.

Discounted Venice-backed raw inference route for `claude-opus-4-6` through the TradeOS AntSeed/OpenAI-compatible surface. This is model output only, not a TradeOS evidence, signal, risk, market, order-flow, or dataset claim.

Discounted Venice-backed raw inference route for `claude-opus-4-7` through the TradeOS AntSeed/OpenAI-compatible surface. This is model output only, not a TradeOS evidence, signal, risk, market, order-flow, or dataset claim.

Discounted Venice-backed raw inference route for `claude-opus-4-8` through the TradeOS AntSeed/OpenAI-compatible surface. This is model output only, not a TradeOS evidence, signal, risk, market, order-flow, or dataset claim.

Discounted Venice-backed raw inference route for `claude-opus-4-8-fast` through the TradeOS AntSeed/OpenAI-compatible surface. This is model output only, not a TradeOS evidence, signal, risk, market, order-flow, or dataset claim.

Routes a natural-language builder or agent request to the most relevant TradeOS intelligence SKU, builds a bounded TradeOS evidence packet, and uses Venice as the explanation layer. The result is a model-facing read with routed SKU, evidence freshness, caveats, and next TradeOS calls. This is not raw inference, execution, custody, copy trading, or personalized financial advice.

Generic Venice model inference exposed through the TradeOS AntSeed/OpenAI-compatible surface. This SKU is useful for agent builders that want a model response without TradeOS evidence grounding. It is explicitly not a TradeOS signal, risk, market, order-flow, or dataset claim; use the Venice Intel Router or direct TradeOS SKUs when provenance, freshness, and evidence matter.

Check order-flow and crash-pressure stress before a bot or analyst acts. Returns active stress symbols, stress levels, how fast stress is changing, crash-pressure level, alert level, VPIN flow-risk v2 state, time-to-stress context, and recent VPIN shock incident evidence showing first seen, later low, drawdown, and minutes to low.