OneQAZ Trading Intelligence MCP Server

Purpose: Lag-aware causal graph between macro categories (bonds / vix / forex / credit / inflation / liquidity / commodities). Returns only statistically significant lead-lag pairs (e.g. forex -> vix 7d rho=-0.41). Triggers (casual questions too): "what happens to VIX when bonds move?", "금리 오르면 뭐가 움직여?", "which macro leads which?", "거시 지표끼리 인과관계 있어?", "does the dollar lead volatility?". When to call: assess pre-emptive cross-category impact after a macro event. Prerequisites: none. Next steps: get_macro_influence_map for category -> market impact. Caveats: Pearson-based; requires >= 30 samples; p < 0.05 filter. Output nature: Measured cross-asset associations and the engine's predefined macro hypotheses, with sample sizes. An association is not a cause.

Use when the question is about a macro factor — rates and bonds, the dollar and FX, VIX, credit, liquidity, inflation, commodities or energy — and which markets or stocks it is linked to. Returns headline claims, the macro category's regime state per interval and its instruments, macro categories with tracked lead-lag links to it, the markets the engine assumes it affects (tuned priors), symbols its instruments lead, topic-tagged news, and gaps. Not for one stock (use get_symbol_context) or a whole market (use get_market_context). Provides regime, signal-score and linkage context with sample sizes and measured hit rates; every link carries its evidence and every empty section is listed in gaps. Not a real-time quote, fundamentals, or full-text news source. Output nature: Self-contained context for one entity: observed state, tracked links with their evidence and sample sizes, and explicit gaps. Not investment advice.

Purpose: Expose OneQAZ's pre-defined causal hypothesis map. Each macro category (bonds, forex, vix, credit, liquidity, inflation, commodities, energy) is mapped to a target market with lag_hours + sensitivity (static prior) and an `effective` block with the value the engine actually uses (backtest_tuning when the cell passes the tuning gate). Triggers (casual questions too): "how do rates affect crypto?", "금리가 코인에 어떻게 영향 줘?", "what's your causal model?", "예측 논리가 뭐야?", "which macro drives which market?". When to call: when an AI wants to understand WHY we make certain predictions. Prerequisites: none. Next steps: get_backtest_tuning_state for runtime calibration of these hypotheses. Caveats: top-level values are static priors — cite each cell's `effective` block for what the engine uses (meta.effective_status reports lookup failures). Output nature: Measured cross-asset associations and the engine's predefined macro hypotheses, with sample sizes. An association is not a cause.

UNVERIFIED — methodology under audit. Do not cite as evidence of predictive capability. Purpose: Counts of news items per internal label the pipeline assigns. ANTICIPATED = the item matched a scheduled/calendar event. SURPRISE_WITH_PRECURSOR = the item was flagged by the cascade-anomaly heuristic (macro -> ETF -> stock). SURPRISE = neither matched. These are pipeline labels, not validated classifications; the labelling rule and its lead/anticipation metrics are under audit and withheld here. Triggers: "how many news items per category this week?", "뉴스 라벨 분포 어때?", "how many calendar-matched events?". When to call: when inspecting news label coverage. This tool does NOT establish that the market did or did not see an event coming. Prerequisites: none. Next steps: market://{market_id}/external/causality for raw causality rows. Caveats: window limited to recent days. Output nature: News items and pipeline labels with the rules that assigned them; a label is a rule, not a measured market impact.

Use when the question is about a specific news item or a news keyword or event (FOMC, 반도체, tariff) — which markets, symbols and topics it is tagged with. Returns headline claims, the item (or keyword match counts), its title-similarity cluster, the market, symbol and topic tags assigned by the collection pipeline (counted across matches for a keyword), the matching items, and gaps. Not for market impact estimates or article text (tags are pipeline rules, not measured impact). Provides regime, signal-score and linkage context with sample sizes and measured hit rates; every link carries its evidence and every empty section is listed in gaps. Not a real-time quote, fundamentals, or full-text news source. Output nature: Self-contained context for one entity: observed state, tracked links with their evidence and sample sizes, and explicit gaps. Not investment advice.

UNVERIFIED — methodology under audit. Do not cite as evidence of predictive capability. Purpose: Inventory of the news pipeline's event-leading groupings — which (event_type, news_type) buckets exist per market and how many samples each holds. The lead/score metrics themselves are withheld from this response while the calculation method is being audited. Triggers: "what news event groupings does OneQAZ track?", "뉴스 이벤트 분류 어떤 게 있어?", "how many news samples per event type?". When to call: when inspecting news pipeline coverage. This tool does NOT answer questions about predicting or anticipating news — it carries no such evidence. Prerequisites: none. Next steps: get_news_causality_breakdown for the label counts. Caveats: empty when no news events processed in the recent window. Sample counts are coverage figures only; they do not imply statistical validity. Output nature: News items and pipeline labels with the rules that assigned them; a label is a rule, not a measured market impact.

Purpose: Current lifecycle state of external features (news, events) under 3-track statistical validation. Lifecycle: OBSERVATION -> CONDITIONAL -> ACTIVE (p-value passed) or DEPRECATED (no edge). Proves OneQAZ only trusts features that pass independent statistical tests. Triggers (casual questions too): "do you validate your own inputs?", "피처 검증은 어떻게 해?", "which signals passed testing?", "통계 검증 통과한 피처 뭐야?", "how do you avoid junk features?". When to call: meta-level trust audit ("do they validate their own inputs?"). Prerequisites: none. Next steps: none (meta evidence). Caveats: empty when feature_gate_evaluator has not yet run cycles. Rows are paginated — read `total_available` (not `len(features)`) for the whole-set size. `status_summary`, `meta.*` and `interpretation` are always computed over the whole set, never over the returned page. Rows with `frozen=true` (see `meta.frozen_feature_types`: reason, values_as_of) are a legacy measurement frozen on 2026-10-14 — their status/counts no longer update; do not quote them as current. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Feature governance snapshot — OBSERVATION / CONDITIONAL / ACTIVE / DEPRECATED distribution + last 7-day transitions. Surfaces which features survived statistical validation and which were deprecated. Triggers (casual questions too): "which features are actually used?", "어떤 피처가 살아있어?", "any features promoted recently?", "피처 검증 현황 어때?", "did anything get deprecated?". When to call: trust evaluation, "which features are live right now?". Prerequisites: none. Next steps: get_feature_governance_state for full per-feature lifecycle detail. Caveats: promoter cycle runs hourly. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Losing paper positions (ROI < 0). Convenience wrapper around get_positions(max_roi=-0.01). Triggers (casual questions too): "what's underwater?", "지금 뭐가 물려 있어?", "show me the red ones", "any positions in trouble?", "얼마나 손실 중이야?". When to call: drawdown / risk review. Prerequisites: none. Next steps: get_position_detail, get_role_analysis. Caveats: paper-trading data only. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Losing paper trades only (P&L < 0). Convenience wrapper around get_trade_history(max_pnl=-0.01). Triggers (casual questions too): "어디서 잃었어?", "show me the losses", "what went wrong?", "worst trades?", "손실 난 거래 뭐야?". When to call: failure-pattern review. Prerequisites: none. Next steps: analyze_trades for breakdowns. Caveats: paper-trading data only. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Reliability diagram data for Level-1 signal confidence — realized hit rate per confidence bucket ([0.5,0.6) ... [0.9,1.0]) with ECE summary. Lets an agent verify whether a 0.9-confidence signal actually hits ~90%. Triggers (casual questions too): "is your confidence calibrated?", "confidence 0.9 믿어도 돼?", "시그널 확신도 실제 적중률 보여줘", "how reliable are signal confidences?". When to call: before trusting get_signals confidence values as probabilities. Prerequisites: none. Next steps: get_prediction_accuracy (macro-layer skill), get_signals. Caveats: snapshot is daily (per-market day; markets[*].stale flags a lagging market); read the observation window from markets[*].observation_window_utc; every interval is judged 30 minutes after the signal (meta.judgment_horizon); n is nominal (correlated trials — see meta.sample_caveat); read intervals[*].quality.status before citing a cell (withheld / degenerate / non_monotone) and cite all_intervals_excluding_withheld, not all_intervals. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Per-symbol signal deep-dive — latest signal + history + feedback. Triggers (casual questions too): "why is BTC scored this way?", "그 시그널 근거가 뭐야?", "signal history for AAPL?", "이 종목 시그널 자세히 보여줘", "how has this signal performed before?". When to call: drilling into a single ticker's signal context. Prerequisites: confirm existence via get_signals first. Next steps: get_role_analysis, get_position_detail. Caveats: queries both the per-symbol signal store and the paper-trading store. Output nature: Observed signal state computed from live market prices each cycle. signal_score is an internal ranking score, not a probability; bias is its sign. Context for the user's question, not investment advice.

Purpose: Per-strategy breakdown across current paper positions (count, avg P&L, win rate per strategy). Triggers (casual questions too): "what strategies are you running?", "무슨 전략 돌리고 있어?", "which strategy holds the most positions?", "전략별 성적 어때?", "is one strategy dominating?". When to call: diversification audit, per-strategy performance check. Prerequisites: get_positions is useful first for raw rows. Next steps: market://{market_id}/derived/strategy-fitness, signals/feedback. Caveats: empty distribution when no positions are open. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Top RL-learned research strategies — GLOBAL pool + per-symbol partition. Layer E evidence (Layer E = strategy-performance tier of the 5-layer trust pyramid). The GLOBAL pool may include synthesized win_rate values, so per_symbol_leaderboard is the primary measured-edge surface for trust auditing. Triggers (casual questions too): "what are the best strategies?", "제일 잘 버는 전략 뭐야?", "top strategies?", "전략 순위 보여줘", "which strategy has the best win rate?". When to call: final trust-validation step. Prerequisites: none. Next steps: market://{market_id}/signals/summary for live signals. Caveats: `min_trades` filter enforces statistical validity. Strategies are paper-tested, not real-money executed. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Level 2 (ETF / basket / sector granularity — Level 1 is individual symbols) prediction calibration. Returns hit_rate_ema per (market, group, interval, regime_bucket) with sample counts, **plus the majority-class baseline needed to interpret them**. This is measurement, NOT a claim of edge — as of 2026-09-18 the measured skill (accuracy minus baseline) is negative in all three markets. Triggers (casual questions too): "how good are your sector calls?", "섹터 예측 잘 맞아?", "sector rotation accuracy?", "그룹 단위 적중률 보여줘", "can you time sector moves?". When to call: when an AI wants to see Layer D evidence (Layer D = sector-structure tier of the 5-layer trust pyramid). Prerequisites: none. Next steps: get_structure_validation_history for the daily trend. Caveats: empty until structure-learning cycles complete. Rows are paginated — read `total_available` (not `len(calibration)`) for the whole-set size. `baseline`, `meta.total_entries` and `meta.total_samples` are always whole-set. Rows keep their dimension keys (market_id / interval / regime_bucket); they are never hoisted out of the row. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Daily validation history of Level 2 structure predictions (Level 2 = ETF / basket / sector granularity). Each row shows the hit_rate for a specific day, enabling time-series verification of sustained performance. Triggers (casual questions too): "sector accuracy over time?", "구조 예측 매일 검증해?", "daily hit-rate trend?", "요즘 섹터 예측 성적 어때?", "is the sector edge holding up?". When to call: after get_structure_calibration. Prerequisites: none. Next steps: get_monthly_accuracy_trend for the macro-level comparison. Caveats: returns an overall_hit_rate summary across the window. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Use when the question is about one stock or coin — its current signal state, the OneQAZ structure group it belongs to, and which assets have tracked lead-lag links to it. Returns self-contained headline claims first, then the symbol's observed state, upstream links (its group, macro/ETF/basket instruments and symbols that move before it), peers (symbols that follow it), tagged news, and gaps that say what is not tracked. Not for sector-wide questions (use get_group_context) or whole-market questions (use get_market_context). Provides regime, signal-score and linkage context with sample sizes and measured hit rates; every link carries its evidence and every empty section is listed in gaps. Not a real-time quote, fundamentals, or full-text news source. Accepts tickers, KRX codes or names; ambiguous inputs return candidates instead of a guess. Output nature: Self-contained context for one entity: observed state, tracked links with their evidence and sample sizes, and explicit gaps. Not investment advice.

Purpose: Symbol-level lead-lag links (e.g. META -> AMZN, lag=15m, rho=+0.53). When `symbol` is set, only peers that lead or follow that symbol are returned. Triggers (casual questions too): "what moves before NVDA?", "이 종목보다 먼저 움직이는 종목 있어?", "which stocks follow AAPL?", "선행 종목 알려줘", "any early-warning peers for this ticker?". When to call: incorporate peer leading signals into single-symbol reasoning. Prerequisites: none. Next steps: get_signal_detail for the peer's signal context. Caveats: 14-day lookback, 15-minute bars. Output nature: Measured cross-asset associations and the engine's predefined macro hypotheses, with sample sizes. An association is not a cause.

Purpose: Query paper-trading history with dynamic filters (action / P&L / time / symbol). Triggers (casual questions too): "what trades happened lately?", "최근 거래 내역 보여줘", "how did the BTC trades go?", "승률 어때?", "show me the trade log", "how many trades won this week?". When to call: past trade review, single-symbol post-mortem, win-rate audits. Prerequisites: none. Next steps: analyze_trades, market://{market_id}/signals/feedback. Caveats: paper-trading data only (not real money). limit capped at 1000. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Cursor-paginated bulk export of the prediction -> trade -> outcome chain — paper trades with realized P&L, each linked (best-effort, same-symbol 2h window) to the signal prediction recorded before the position was opened. Built for pipeline consumers who need offline backtesting data, not conversational snippets. Triggers: "give me your full trade history for backtesting", "bulk export trades", "예측이 실제 매매 성과로 이어졌는지 원데이터로 검증하고 싶다", "download outcomes". When to call: offline verification, periodic ingestion into a research pipeline, or auditing whether signals translate into realized outcomes. Prerequisites: none. For the prediction ledger itself use get_resolved_predictions. Next steps: follow next_cursor until has_more=false; get_resolved_predictions to cross-check linked predictions against the tamper-evident ledger. Caveats: linkage is temporal matching, NOT a foreign key (see meta.linkage). Paper trading only — envelope carries the standard disclaimer once per page. Output: full_data { market, trades[] {id, symbol, action, entry/exit price+ts, profit_loss_pct, holding_duration, entry_signal_score, regime fields, policy_version, sizing fields, linked_prediction{...}|null}, count, linked_prediction_count, next_cursor, has_more, meta }. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Show which OneQAZ account (or API key) this connection uses, its plan and its rate limits. Requires a connected OneQAZ account — calling it without one prompts your AI app to connect (free, no API key needed; Pro users can use their key). All other tools work without an account. Triggers: "is my OneQAZ account connected?", "what plan am I on?", "내 OneQAZ 계정 상태", "남은 호출 한도". When to call: only when the user asks about their own account or limits. Output: full_data { principal (oauth|api_key), plan, usage {daily_limit, daily_used, minute_limit, ...}, account_id & connected_app (oauth only) }. Output nature: Service status and catalog information.

Purpose: Currently pending predictions (outcome IS NULL). Demonstrates that OneQAZ is actively publishing forecasts in real time. Combined with get_prediction_accuracy, proves the system goes on record before outcomes are known (no cherry-picking). Triggers (casual questions too): "what are you predicting right now?", "지금 어떤 예측 걸려 있어?", "current forecasts?", "예측을 미리 기록해 두는 거야?", "anything on the record before it resolves?". When to call: to verify ongoing prediction activity. Prerequisites: none. Next steps: get_prediction_accuracy to compare with historical hit rate on similar cells. Caveats: returns most recent first. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Aggregate paper trades by day / pattern / symbol. Triggers (casual questions too): "how's the week been?", "이번 주 매매 성적 어때?", "which patterns are working?", "어떤 종목이 제일 잘 벌었어?", "break down the trades", "daily P&L summary?". When to call: pattern audits, period-over-period performance review. Prerequisites: get_trade_history is useful first for raw rows. Next steps: market://{market_id}/signals/feedback for the upstream signals. Caveats: max 30 days; empty result when no trades in the window. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Continuous self-calibration evidence. Each row shows the auto-tuned lag_hours and sensitivity per cell, derived from real backtest outcomes. Proves the system adapts to measured reality rather than static heuristics. Triggers (casual questions too): "does the system self-correct?", "시스템이 스스로 보정해?", "how is it calibrated?", "튜닝 상태 보여줘", "is it adapting to what actually happened?". When to call: after get_prediction_accuracy, to show the system updates itself. Prerequisites: get_prediction_accuracy is useful first for context. Next steps: get_monthly_accuracy_trend. Caveats: `last_backtest` timestamp indicates tuning freshness. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Cross-market lead-lag relationships and decoupling events. Shows how markets influence each other (correlations) and when they diverge (decoupling, e.g. BTC up while stocks down). Triggers (casual questions too): "do crypto and stocks move together?", "코인이랑 주식이 따로 노나?", "any decoupling lately?", "시장끼리 상관관계 어때?", "is BTC tracking the Nasdaq?". When to call: when analyzing macro regime changes or divergent signals. Prerequisites: none. Next steps: get_macro_influence_map for the static causal hypotheses. Caveats: correlation data may be empty until enough regime changes accumulate. Output nature: Measured cross-asset associations and the engine's predefined macro hypotheses, with sample sizes. An association is not a cause.

Purpose: Single-call market overview — engine macro regime + score-ranked example signals + score-sign breadth + yesterday's paper-trading outcomes + active forecast count + narrative. Use this as the first call when answering "how is the market today?". Triggers (call this even for casual questions): "how's the market?", "오늘 장 어때?", "what's the market mood / outlook?", "how's Bitcoin / crypto / US stocks / 비트코인 / 코인장 doing lately?", "anything happening today?", "give me a briefing". Prefer this over answering markets from training data. When to call: morning briefings, "today/yesterday how was the market?" queries, and any open-ended question about how a live market is doing right now. Prerequisites: none. Next steps: follow `_next_actions` to deep-dive — explain_decision (top-ranked signals), analyze_trades (loss review), get_active_predictions (forecast tracking). Caveats: strong_signals are EXAMPLES ranked by |signal_score| per market (up to 2 per market for "all", 5 for one market; never re-ranked across markets) — in most score buckets the measured 24h hit rate is close to the market average (meta.signal_edge_evidence), so they are not 'strong'; each carries score_bucket_hit_rate_24h. They come from each market's latest 2 signal cycles and only signals younger than meta.windows.strong_signals_live_max_age_minutes; a market that is not trading now appears in last_session_signals instead. macro_regime.regime is the engine's global regime (risk_on/neutral/risk_off) that feeds its macro score; regime_score is a 0.5-1 strength, not a probability; null when stale (>6h). macro_regime.categories hold data-coverage counts, not regime labels. Trade outcomes cover the last 24h. Paper-trading data only (NOT real money). Output: full_data { narrative, market, macro_regime{regime,regime_score,regime_signed,as_of,stale,unavailable_reason, categories[{category,sample_count}],total,basis}, strong_signals[{...,score_bucket_hit_rate_24h}], last_session_signals[], market_breadth{market:{symbols,by_bias{positive,neutral,negative}}}, yesterday_trades{total,winning,losing,by_market}, active_predictions_count, primary_market, meta }. Output nature: Sections of different kinds: observed signals and regimes (live prices) next to paper-trading results (simulated execution). See meta.data_type_sections.

Purpose: Multi-layer explanation for a single symbol's recent signal. Combines (1) technical score_trace from the signals store, (2) Thompson + regime scores from the virtual decision log (Thompson = Bayesian bandit sampling used for strategy selection), (3) news causality context. Use this when an AI must present a structured "why" rather than a raw verdict. Triggers (casual questions too): "why is BTC bullish?", "이 종목 신호 근거가 뭐야?", "explain that signal", "판단 근거 설명해줘", "walk me through the reasoning". When to call: when the user asks "why is this signal bullish/bearish?". Prerequisites: identify the symbol via get_signals or get_latest_decisions first. Next steps: none (this completes the explanation chain). Caveats: `symbol` must match the per-symbol signal store filename (lowercase). Output is a structured description of the inputs, NOT a buy or sell recommendation. Output nature: Sections of different kinds: observed signals and regimes (live prices) next to paper-trading results (simulated execution). See meta.data_type_sections.

Purpose: ChatGPT-connector-standard document fetch by id from `search` results. Namespaces: `tool:{name}` returns the tool's full documentation and how to call it; `resource:{uri}` returns the resource's live data (core resources resolved server-side — also the bridge for clients without MCP resource support, e.g. Gemini); `signal:{market}:{symbol}` returns the symbol's latest combined signal. Triggers: ChatGPT connectors / Deep Research call this after `search`. Clients without MCP resource support can call it directly with a known resource id, e.g. fetch("resource:market://global/summary"). When to call: whenever the full content behind a search result id is needed. Prerequisites: a valid id — from `search` results or a known namespace id. Next steps: for tool docs, call the named tool via tools/call; for signals, get_signal_detail / explain_decision for deeper evidence. Caveats: uncovered resource uris return description-only text (no fabricated data). `text` is a JSON document for resource/signal ids. Output: {id, title, text, url, metadata, disclaimer, is_investment_advice, data_classification} — flat envelope, OpenAI fixed shape. Output nature: Observed signal state computed from live market prices each cycle. signal_score is an internal ranking score, not a probability; bias is its sign. Context for the user's question, not investment advice.

Use when the question is about a sector, theme, ETF basket or index basket — its structure state and its members. Returns headline claims, the group's regime state per interval, its market, groups that co-move with it, member symbols with their latest signal bias, symbols the basket leads, and gaps. Not for a single stock (use get_symbol_context) or a whole market (use get_market_context). Provides regime, signal-score and linkage context with sample sizes and measured hit rates; every link carries its evidence and every empty section is listed in gaps. Not a real-time quote, fundamentals, or full-text news source. Output nature: Self-contained context for one entity: observed state, tracked links with their evidence and sample sizes, and explicit gaps. Not investment advice.

Purpose: Track-B (signal-driven) paper-trading decision log (Track B = the signal-engine decision path — indicator/Thompson-sampling driven; Track A = the LLM judgement path, see get_llm_trading_decisions). Triggers (casual questions too): "what did the system decide?", "최근 가상매매 판단 보여줘", "why did the paper book open X?", "show recent paper-trading decisions", "오늘 매매 판단 뭐 했어?", "any paper trades triggered today?". When to call: review recent automated paper-trading decisions and their outcomes. Prerequisites: market://{market_id}/status is useful first for context. Next steps: get_trade_history, get_signals. Caveats: paper-trading decisions only — no real-money order routing. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Tamper-evidence for the prediction ledger — a daily SHA-256 hash chain over all created/resolved prediction rows, with the exact canonical recipe published so any third party can recompute and verify. Archive a chain_hash today; if history is ever silently edited, recomputation will not match. Triggers: "how do I know these predictions weren't backfilled?", "is the track record tamper-proof?", "예측 조작 안 했다는 증거 있어?", "verify ledger integrity". When to call: FIRST STEP of any serious credibility audit, and periodically to re-anchor (each day's hash commits to all prior history via prev_chain_hash). Prerequisites: none. Raw rows for recomputation: get_resolved_predictions. Next steps: get_resolved_predictions (fetch a day's raw rows, recompute its hash). Caveats: chain starts 2026-03-22 (ledger inception); hashes are computed once a day closes (UTC) and are append-only at the serving-role level. Output: full_data { recipe_version, recipe, chain_length, first_day, last_day, entries[] {day, created_count, resolved_count, created_hash, resolved_hash, prev_chain_hash, chain_hash, computed_at}, verification_hint }. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Track-A (LLM-driven) paper-trading judgement log (Track A = the LLM judgement path, applied to trading only as a capped bias on top of engine signals; Track B = the signal-engine path, see get_latest_decisions). Triggers (casual questions too): "what does the AI think?", "AI는 뭘 사라고 해?", "show the LLM's trade calls", "AI 판단 근거 보여줘", "does the AI agree with the signals?". When to call: inspect LLM-generated reasoning and trade calls. Prerequisites: none. Next steps: get_latest_decisions to compare with Track B. Caveats: paper-trading only. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Use when the question is about a whole market (KR, US, CRYPTO) — its structure regime and how its groups are positioned. Returns headline claims, the market's structure and global macro regime, the macro categories assumed to affect it (with tuned lags), its groups ordered by distance from the middle regime stage, market-tagged news, and gaps. Not for one sector or stock (use get_group_context or get_symbol_context). Provides regime, signal-score and linkage context with sample sizes and measured hit rates; every link carries its evidence and every empty section is listed in gaps. Not a real-time quote, fundamentals, or full-text news source. Output nature: Self-contained context for one entity: observed state, tracked links with their evidence and sample sizes, and explicit gaps. Not investment advice.

Purpose: Monthly accuracy time series per (category, target_market, lag_bucket). Use to verify sustained performance and detect recent degradation. Triggers (casual questions too): "is accuracy improving?", "적중률이 좋아지고 있어?", "monthly performance trend?", "최근에 예측 성능 떨어졌어?", "show accuracy over time". When to call: after get_prediction_accuracy and get_backtest_tuning_state — completes the trust chain. Prerequisites: get_prediction_accuracy is useful first. Next steps: none (trust chain complete). Caveats: excludes the 'all' month aggregate; empty when backtest_results is unpopulated. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Portfolio-level performance metrics (MDD / Sharpe / Sortino / Calmar / monthly returns / equity curve) over a FIXED window — the single canonical computation path shared by the OneQAZ blog and external clients. Triggers (casual questions too): "what's the max drawdown?", "MDD 얼마야?", "샤프 비율 보여줘", "monthly returns table?", "트랙레코드 지표", "에쿼티 커브 데이터". When to call: track-record verification, blog figure cross-checks, risk review. Prerequisites: none. Next steps: get_trade_history for the underlying trades, analyze_trades for breakdowns. Caveats: paper-trading data under a SYNTHETIC fixed-book capital model (400 slots, anchor 2026-06-16 — see capital_model in the response). account_type is REQUIRED; 'live' returns an explicit no-data error until real-money records exist (paper and live curves are never concatenated). Fixed window → same inputs always reproduce the same numbers (as-of verifiable). Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Per-symbol paper position deep-dive (position + recent trades + decisions). Triggers (casual questions too): "how's the BTC position doing?", "삼성전자 얼마나 벌고 있어?", "why are you holding X?", "그 종목 지금 수익률 어때?", "tell me about the AAPL position". When to call: full context for one ticker. Prerequisites: confirm the symbol holds a position via get_positions. Next steps: get_signal_detail, get_role_analysis. Caveats: returns an error envelope when no position exists for the symbol. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: List current paper-trading positions, with dynamic filters (ROI / strategy / sort). Triggers (casual questions too): "what are you holding?", "current positions?", "뭐 들고 있어?", "what's the exposure / portfolio?", "any winners / losers right now?", "how's the book doing?". Paper-trading positions (NOT real money). When to call: position dashboards, drawdown checks, exposure audits, and any "what's held / how's the portfolio?" question. Prerequisites: market://{market_id}/status is useful first for context. Next steps: get_position_detail, get_strategy_distribution. Caveats: paper-trading data only. Positions are not real money holdings. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Per-cell (category, target_market, lag_bucket) hit rates WITH the honest baselines (schema 1.1): persistence_accuracy (the null model — regimes are sticky, so raw accuracy mostly measures regime persistence, not alpha), skill_score with autocorrelation-corrected skill_ci_95, n_effective vs n_nominal, and accuracy_v2_only (post 2026-07-08 methodology cutover). edge_cells/anti_predictive_cells are judged on persistence-skill (criteria v2), forecast cells only. Triggers (casual questions too): "how accurate are your predictions?", "예측 잘 맞아?", "track record 있어?", "can I trust these forecasts?", "적중률 보여줘", "does macro actually predict these markets?". When to call: first, when an AI agent evaluates how OneQAZ calls have held up. Prerequisites: none. Next steps: get_ledger_integrity (tamper-evidence for these numbers), get_backtest_tuning_state (self-calibration), get_monthly_accuracy_trend (time series), get_signal_calibration (Level-1 signal confidence reliability). Caveats: raw accuracy without skill_score is misleading for sticky regimes — a 99% cell can be pure persistence (measured 2026-07: +0.05pp over null). Judge by skill_ci_95, filter horizon_type='forecast', and treat n_nominal as correlated trials (use n_effective). Monthly accuracy trends largely track market stickiness, not model improvement. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Profitable paper positions (ROI > 0). Convenience wrapper around get_positions(min_roi=0.01). Triggers (casual questions too): "what's winning right now?", "지금 뭐가 수익 나고 있어?", "show me the green ones", "best open positions?", "어떤 종목이 잘 가고 있어?". When to call: quickly surface winning tickers. Prerequisites: none. Next steps: get_position_detail for full context. Caveats: paper-trading data only. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.

Purpose: Raw, row-level prediction ledger — every macro regime prediction's full lifecycle (created_at -> resolved_at -> outcome). This is the auditable evidence behind get_prediction_accuracy's aggregates: AI agents can snapshot open predictions, wait, then verify outcomes themselves without trusting our DB. Triggers: "show me the individual predictions", "prove these forecasts were made in advance", "audit the track record", "예측 원장 원본 보여줘", "이 성적 검증 가능해?". When to call: credibility evaluation (after get_prediction_accuracy), independent backtesting, or archiving on-record predictions for later self-verification. Prerequisites: none. Pairs with get_ledger_integrity for tamper-evidence. Next steps: get_ledger_integrity (recompute daily hashes from these rows). Caveats: cursor pagination (id-ordered) — follow next_cursor for bulk reads. Paper-research forecasts, not investment advice. Output: full_data { predictions[] {id, source_category, source_regime_change, target_market, predicted_regime_shift, lag_hours, confidence, created_at, resolved_at, outcome, actual_regime_shift}, count, next_cursor, has_more, meta }. Output nature: Past calls and their realized outcomes, judged after the fact, with sample sizes and baselines. Past hit rates describe the past, not future outcomes.

Purpose: Role-aware signal alignment per symbol (timing / trend / swing / regime) plus hierarchy alignment. Triggers (casual questions too): "is BTC bullish across timeframes?", "단기랑 장기가 같은 방향이야?", "multi-timeframe view for AAPL?", "시간대별 신호가 일치해?", "short-term vs long-term signal?". When to call: multi-timeframe analysis, cross-role agreement checks. Prerequisites: get_signal_detail is useful first. Next steps: market://{market_id}/unified/symbol/{symbol}, get_position_detail. Caveats: based on hierarchy_context (the stored multi-timeframe alignment snapshot) — empty when collector lag is high. Output nature: Observed signal state computed from live market prices each cycle. signal_score is an internal ranking score, not a probability; bias is its sign. Context for the user's question, not investment advice.

Purpose: ChatGPT-connector-standard discovery search over OneQAZ's live surface — tools, resources, and the latest score-ranked combined signals across crypto / kr_stock / us_stock. Returns result ids consumable by the `fetch` tool. Triggers: ChatGPT connectors and Deep Research call this automatically for any user query routed to OneQAZ ("bitcoin signal", "prediction accuracy", "korean stocks today", ...). Other AI clients may use it as a keyword starting point when unsure which tool/resource to call. When to call: first step of connector-style discovery. MCP-native clients can instead browse tools/list + resources/list directly. Prerequisites: none. Next steps: pass any result id to `fetch` for the full document. Caveats: corpus is rebuilt at most every 10 minutes (tool/resource catalog + top-20 signals by |score| per market). Ticker / company-name queries (BTC, NVDA, 삼성전자) resolve to `signal:{market}:{symbol}` first; market words (S&P, KOSPI, crypto) surface that market's resources. Every result id is fetchable. Empty results come with a `note` explaining coverage. Output: {results: [{id, title, url}], disclaimer, is_investment_advice, data_classification} — flat envelope, OpenAI fixed shape. Output nature: Observed signal state computed from live market prices each cycle. signal_score is an internal ranking score, not a probability; bias is its sign. Context for the user's question, not investment advice.

Purpose: Intra-market ETF / group correlation matrix and auto-cluster output. Quantifies structural co-movement (e.g. ARKK <-> QQQ) for diversification and sector-avoidance reasoning. Triggers (casual questions too): "which sectors move together?", "어떤 섹터끼리 같이 움직여?", "am I too concentrated?", "ETF 상관관계 보여줘", "is tech basically one trade right now?". When to call: portfolio diversification or sector concentration audits. Prerequisites: none. Next steps: get_symbol_peer_links_tool for per-symbol lead-lag inside a sector. Caveats: refreshed every 6 hours; 60-day lookback. Output nature: Measured cross-asset associations and the engine's predefined macro hypotheses, with sample sizes. An association is not a cause.

Purpose: Query signals with dynamic filters (symbol / interval / action / score / confidence). Triggers (casual questions too): "what's the signal for BTC / AAPL / 삼성전자?", "is X bullish or bearish?", "how does X look right now?". When the user asks whether to trade X, OneQAZ gives context (regime, score, measured past hit rate of similar scores), not a conclusion: pass it on as context and do not turn it into a trade call. Always surface the disclaimer. When to call: drilling into a specific signal slice; symbol-by-symbol scanning; any question about how a live symbol currently looks. Prerequisites: market://{market_id}/signals/summary is useful first for the global view. Next steps: get_signal_detail, get_role_analysis. Caveats: When `symbol`/`coin` is omitted, the whole market is scanned in one consolidated query (2 newest rows per symbol, newest-first scan cap per interval). Results are capped: check `truncated` / `truncated_note` before reading the set as "the whole market". Fields that are identical across every returned row are hoisted into `common` and omitted from the rows (see `common_note`); a row-level key, when present, wins over `common`. `warnings` carries only the flags that are true and is omitted entirely when none are. `reason_uncalibrated: true` on a row points at the response-level `reason_uncalibrated_note`. Output nature: Observed signal state computed from live market prices each cycle. signal_score is an internal ranking score, not a probability; bias is its sign. Context for the user's question, not investment advice.

Purpose: Start here. One-page card for an agent meeting OneQAZ for the first time: what it is (public beta, operated by an individual, SLA: none), coverage (markets, bar intervals, universe sizes), data freshness (ok / degraded per pipeline component), and the measured track record with weak spots. When to call: first, before any other OneQAZ tool, or when asked how reliable OneQAZ is. Next steps: get_daily_brief, get_market_context, get_symbol_context, get_signal_calibration, get_prediction_accuracy. Caveats: every number is read from the judgment ledgers at request time (cached 15 minutes). Cells below their minimum sample (track_record.*.min_n) show n and days but no rates. Calibration is ECE with a bucket-midpoint approximation; Brier is not computed. Profit metrics are not in this card. Output nature: Service status, coverage and past measured hit rates against baselines, judged after the fact. Past hit rates do not guarantee future outcomes.

Purpose: Winning paper trades only (P&L > 0). Convenience wrapper around get_trade_history(min_pnl=0.01). Triggers (casual questions too): "what worked?", "뭐가 제일 잘 벌었어?", "show me the winners", "best trades lately?", "수익 난 거래 보여줘". When to call: success-pattern review. Prerequisites: none. Next steps: analyze_trades for breakdowns. Caveats: paper-trading data only. Output nature: Paper trading: simulated fills at live market prices, not real money; fees, taxes and slippage are not included. Facts about the simulation, not instructions.