Data Coverage

At a glance

Cornerstones is a market context layer for AI trading agents. It packages market truth, charts, order flow, events, and context into JSON that customers and agents can verify before using.

What you can read

• Market truth / prices: FX quotes, gold bars, stock quotes, A-share symbols, exchange discovery, symbol normalization, sessions, indicators, bounded universes, and screeners.
• Charts: generated chart artifacts for multimodal inspection, with timeframe, lookback, indicators, template, layout, layers, and request metadata when available.
• Corporate evidence: stock profiles, optionability, earnings schedules, filings, and corporate actions as bounded workflow inputs.
• Events & receipts: recent events, history, receipts, subscriptions, deletes, and evidence references the agent can cite.
• Order flow / market structure: summary, context, raw, historical, and liquidity metrics where account access permits it.
• Context packets: gold, FX, stocks, cross-asset, macro, geopolitics, and prediction-market style context packaged for reasoning loops.
• Account access: start with help and discovery, then use issued keys for Free market truth, Pro charts/context/options, and Max orderflow depth.

How examples are organized

Examples use the same customer workflow every time: ask a market question, run a command, inspect the live response, then confirm whether the account can use that surface.

• Ask: the market question the agent is trying to answer.
• Command: the exact public client command to run.
• Live response: JSON captured from a live Cornerstones API response, then normalized to customer-facing Cornerstones source labels.
• Access: whether the current account can use that surface.
• Response checks: source labels, timestamps, freshness, quality notes, and account access.

Coverage lanes

Market truth

cornerstones-client fx quote --symbol EURUSD
cornerstones-client fx bars --symbol XAUUSD --timeframe 1h --count 3
cornerstones-client stocks quote --symbol AAPL
cornerstones-client stocks quote --symbol 600519.SS
cornerstones-client stocks normalize-symbol --symbol 600519.SH

Use these commands when the agent needs observable market state. Do not turn this into an unlimited export workflow; keep requests bounded to the question.

Agent-visible charts

cornerstones-client chart fx --symbol XAUUSD --timeframe 15m --bars 200 --indicator ema20 --indicator rsi14
cornerstones-client chart stocks --symbol AAPL --timeframe 1d --bars 120 --indicator sma50

Charts are evidence objects, not decorative screenshots. A multimodal agent should inspect the artifact and then check the manifest, freshness, and source labels.

Order flow and structure

cornerstones-client orderflow summary --symbol XAUUSD
cornerstones-client orderflow context --symbol XAUUSD
cornerstones-client orderflow liquidity-metrics --symbol XAUUSD

Order flow is premium market-structure context. Treat unavailable states as facts about the account, not as code failures.

Events and evidence

cornerstones-client events recent
cornerstones-client events history --symbol AAPL
cornerstones-client events receipts
cornerstones-client stocks earnings --symbol AAPL
cornerstones-client stocks filings --symbol AAPL
cornerstones-client stocks corporate-actions --symbol AAPL

Use events as citation material. Carry event time, source label, freshness, and evidence id into downstream reasoning.

Context packets

cornerstones-client context gold --symbol XAUUSD --timeframe 1h --count 3
cornerstones-client context fx --symbol EURUSD --timeframe 1h --count 3
cornerstones-client cross-asset context --symbol XAUUSD

Context packets are the main product metaphor: a compact market state the next agent step can verify before acting.