Capture a screenshot of the TradingView desktop app for visual analysis. By default the whole window (all charts, drawings, watchlist); pass chart_index to capture just one chart of a multi-chart layout at full resolution.

Get the current TradingView layout state: every chart's symbol, timeframe and active indicators as {id, name} (the id is what get_indicator_* tools accept as study_id), plus which chart is active. Call this first to know what the user is looking at.

Read the current TradingView Bar Replay state, active chart binding, historical cursor time, and replay resolution. This is read-only and never starts, advances, or stops replay trading.

Preview or start TradingView Bar Replay on the active chart at one historical instant. expected_symbol and expected_timeframe are checked immediately before the write. confirm=true is required. Replay Trading orders and autoplay are never used.

Advance an already started, paused TradingView Bar Replay by 1-100 bars. Each step is read back and must advance the replay cursor. Autoplay and Replay Trading orders are not supported.

Preview or stop TradingView Bar Replay and return the chart to real-time mode. confirm=true is required when replay or its toolbar is active.

Get OHLCV candle data (time, open, high, low, close, volume) currently loaded in a TradingView chart. Time is a unix timestamp in seconds.

Calibrate one frozen FVG-retest or session-auction event-study decision rule against deterministic, predictability-free synthetic OHLC. It runs each selected null model separately and returns the candidate rate, Wilson interval, seed range, and failures. A candidate requires a global mean confidence interval wholly above zero plus positive evidence in every synthetic fold. It does not read or change TradingView, record a journal entry, rank variants, or establish profitability.

Run a bounded, read-only market event study on closed OHLC bars from the active chart. Condition session_auction classifies the first break of a prior local-session range as accepted outside closes or a failed return inside. Condition session_exhaustion_handoff tests whether a closed-bar prior-session direction fails to extend in an early handoff session and reverses; its outcomes begin only after the configured handoff window has fully closed. Condition event_aftershock_retest evaluates caller-supplied, canonical economic-event timestamps through a post-event initial range, close breakout, and first boundary retest. Condition failed_breakout evaluates a first sweep beyond a completed local-session range that closes back inside it, with optional opposite-direction confirmation closes. It returns directional forward returns, MFE, MAE, target timing, explicit exclusions, and optional non-overlapping time folds with bounded mean and rate confidence intervals. The caller can declare the number of configurations inspected; serial dependence and multiple testing are not silently adjusted. An optional regime split joins each event only to a price/volatility label whose bar closed before the signal bar began, and keeps sparse cells not evaluable. Signal-bar close is an event reference, not an assumed fill. It never ranks parameters, changes the chart, or places orders.

Return the fixed Pine Study template that reads confirmed DXY daily data with lookahead_off and exposes dxy_return_20 and dxy_gate plots. Read-only.

Run a bounded, read-only cross-asset event study over two exact TradingView charts, with an optional third-chart OHLC regime or fixed DXY Pine gate on the target chart. A driver impulse becomes usable only after its nominal bar close; the study then tests whether the target fails to break in the expected direction and confirms an opposite structural close. It uses an as-of join without forward-fill and returns forward return, MFE, MAE, target timing, fold results, and explicit exclusions. Signal-bar close is an event reference, not an assumed fill. It never changes charts, ranks parameters, or places orders.

Calibrate the frozen feature-outcome candidate gate against deterministic synthetic null OHLC. A candidate must pass the horizon-one, non-overlapping Newey-West, Bonferroni, and fixed 1,000-replication empirical-null rule. It does not read or change TradingView, record a journal entry, rank variants, or establish profitability.

Calibrate the frozen lead-lag candidate rule against paired nulls. Factor variants carry contemporaneous dependence but no lagged predictability; independent variants isolate marginal path effects. Every replication runs the rule's own circular-shift empirical null after the v3 default's fixed causal prior-20-return RMS scaling. The result carries its fully resolved configuration and a hash of it, so a quoted rate can be reproduced. It does not read or change TradingView and does not establish profitability.

Measure how often the frozen feature-outcome candidate gate detects a predeclared signed next-bar effect injected after one body-direction bucket. It is synthetic sensitivity evidence, not market alpha.

Measure point-in-time relationships between selected closed-bar price features and later observed returns on one exact TradingView chart. It classifies ATR compression, candle body direction, wick imbalance, directional streaks, range position, and opening gaps using only the signal bar and prior OHLC. It returns bucketed forward-return, upside, downside, and fold distributions without optimizing thresholds, changing the chart, or making a trade recommendation.

Summarize deterministic session-day profiles from closed minute bars on one exact TradingView chart. Sessions use caller-specified IANA timezones and support daylight-saving and cross-midnight boundaries. The tool returns coverage, OHLC range, return, opening-range extension, high/low timing, VWAP, prior closed-session overlap and high/low reaction tests (PDH/PDL/PDC), and TradingView bar volume clearly labelled as unverified tick-or-exchange volume. It does not change the chart, optimize session definitions, or make a trade recommendation.

Classify deterministic directional and volatility regimes from closed OHLC bars already loaded on the active chart. Every label uses only that bar and earlier evidence: efficiency ratio and ATR-normalized directional movement classify trend/range/transition, while current ATR relative to a trailing ATR median classifies low/normal/high volatility. Thresholds are explicit and no future-fitted quantiles, ranking, chart changes, or trade recommendations are used.

Run one exact saved Pine Strategy temporarily, collect its complete immutable trade ledger, and join each closed trade to the latest market-regime bar whose nominal close was available by entry time. Returns PF, expectancy, win rate, closed-trade drawdown, run-up/drawdown, and coverage by directional, volatility, and combined regime. Dry-run by default; confirm=true is required. The strategy is removed and the original chart fingerprint is verified. It never ranks regimes, changes the saved Pine source, or places orders.

Run a bounded serial matrix of exact saved Pine Strategies across explicit symbols and timeframes, then join each complete trade ledger to point-in-time market regimes. Each job reads loaded closed OHLC, temporarily adds one strategy, removes it, and verifies restoration of the original chart. Returns coverage and descriptive performance by regime without ranking or automatic adoption. Dry-run by default; confirm=true is required. It never changes saved Pine source or places orders.

Get recent plot values of indicators (studies) on a TradingView chart — e.g. signal levels, bands, oscillator readings. Plot names come from the indicator's own style titles. Cosmetic plots (colors, alert flags) are excluded by default. Use get_chart_context first to see which indicators exist. Narrow to specific plots with plot_titles, which is required above 500 bars so a long history stays readable.

Get the input parameters (settings) of indicators on a TradingView chart, with names, current values, defaults and tooltips — e.g. 'Pivot Length: 5'.

Change input values of an indicator or strategy already on a chart — the write counterpart to get_indicator_inputs. The Pine source is untouched, but the study's input values on the chart remain changed until set back (this is a live chart edit, like opening the study's Settings dialog, and may be captured by TradingView's own layout autosave). Works for both plain indicators and strategies (for a strategy, follow up with get_strategy_report to read the recalculated backtest). Use this to A/B-test parameters without re-saving the script each time, and restore the original values when done.

Get drawing primitives (labels with text+price, trend lines, boxes/zones) that a Pine indicator has drawn on a TradingView chart. This is how to read drawing-only indicators (e.g. Elliott Wave labels, support/resistance lines, order blocks) that have no numeric plots. Most recent primitives first. Times beyond the last bar are extrapolated and flagged timeEstimated.

Read tables drawn by Pine indicators on a TradingView chart (e.g. a multi-timeframe trend dashboard in the corner) as text grids: grid[row][column] plus the table's on-chart position. This is the only way to read table-only summaries that have no plots or drawings.

Load more historical bars into a TradingView chart (like scrolling left), so that get_ohlcv and get_indicator_values can see further back. The visible chart view is not changed. Returns how many bars were added and the new earliest bar time.

List the user's own saved Pine scripts (indicators and strategies) with their pine_id, kind and version, cross-referenced with the charts: usedBy shows which on-chart indicators are rendered from each script. Use this to find the pine_id for get_pine_source. Read-only.

Get the full Pine source code of one of the user's OWN saved scripts, for review or improvement suggestions. Only 'USER;...' ids from list_pine_scripts are accepted; published/protected third-party scripts are refused. Read-only — editing or saving scripts is not supported.

Get the audited generic Pine overlay used to render a structured market analysis (entry zone, confirmation, invalidation, stop, targets, confidence and expiry). Read-only: pass the returned source to save_pine_script, then add_pine_to_chart once; subsequent analyses should update that study with apply_analysis_overlay.

Get the audited Pine template that exposes only a completed exchange-trading-day chart-bar volume-allocation profile's prior POC, VAH and VAL as readable plots. It does not claim Bid/Ask order flow or institutional activity. Save it with save_pine_script, add it once with add_pine_to_chart, then use get_volume_profile_context.

Read a completed prior-session POC, VAH and VAL only from the exact audited Bushido Volume Profile Context Pine template. The saved source, study placement, symbol and timeframe are verified fail-closed. Results describe a volume-derived price-level context, not direct large-order, Bid/Ask, or order-flow evidence.

Get the audited Pine template that marks pin bars, engulfing bars and 20-bar sweeps, and exposes each as a readable plot. Marks and alerts are gated on bar close, so nothing appears intrabar and then disappears. Save it with save_pine_script, add it once with add_pine_to_chart, then use get_price_action_context.

Read the latest bar's pin-bar, engulfing and sweep readings only from the exact audited Bushido Price Action Context Pine template. The saved source, study placement, symbol and timeframe are verified fail-closed, and so are the study inputs: a study with Confirm On Bar Close switched off is refused, because its signals can change after being read. Any other departure from the audited defaults is reported in settings and flagged in qualityIssues. A signal of 0 on an unconfirmed bar means the bar has not closed yet, not that the pattern is absent.

Measure what the three price-action patterns are worth as naked entries on the chart's own bars: enter at the signal bar's close in the direction the pattern reads, exit h bars later, with no stop, target or filter. Uses the same rules as the Pine template at its default settings. Reports each pattern's forward return with an IID normal-approximation interval for description only alongside the clock hours it fires in and what any bar in those hours does anyway, because a pattern that only fires at a session boundary will otherwise report that boundary as an effect.

Run one frozen descriptive #61b reaction-study variant on CME exchange-volume chart bars and the exact audited Bushido Volume Profile Context proxy. The 240-minute v1 and 60-minute v1 variants are separate methodologies and their evidence must not be combined. Both fix 24 rows, 70% value area, four rejection/acceptance branches, horizons 1/2/4/8 and a 20 bps descriptive target. It also compares each branch with non-event bars in the same prior-closed directional-plus-volatility regime, standardized to the event regime mix. It verifies source, placement, inputs, symbol and timeframe and never ranks or adopts a branch. Native VP, order flow and assumed fills are explicitly out of scope.

Run the separately frozen descriptive #61b 60-minute POC-reversion study on CME exchange-volume chart bars and the exact audited Bushido Volume Profile Context proxy. A signal is the first completed close outside VAH or VAL and at least 20 bps from the completed prior POC; it measures return toward that POC, POC-touch frequency, and a same-prior-regime non-event baseline. This is not the VAH/VAL reaction study and its evidence must not be combined with that population. It is descriptive only and never ranks or adopts a branch.

Save Pine source to the user's script library — the ONLY write tool, with a confirm flow: without confirm=true nothing is written and a dry-run preview is returned; show it to the user and get their approval before calling again with confirm=true. Non-destructive by design: creates a new script (name, no pine_id) or a new version of an existing one (pine_id), and every older version stays retrievable via get_pine_source(pine_id, version). Compile errors are returned with line numbers; note the version is stored even if compilation fails (see revertHint). Typical PDCA loop: get_pine_source -> edit -> save_pine_script -> run_backtest.

Add one of the user's OWN saved Pine scripts (latest version) to a TradingView chart as a study — e.g. to show an improved indicator after save_pine_script. Additive only: never removes or replaces existing studies (the user can remove it from the chart UI). For strategies prefer run_backtest, which cleans up after itself.

Preview or remove one on-chart instance of the user's OWN saved Pine script. The tool verifies the USER pine_id against both list_pine_scripts and the study's hidden Pine id before removal. It also fails closed if the chart symbol or timeframe changed. Without confirm=true nothing is removed.

Idempotently ensure that the audited Bushido Analysis Overlay is present once on a target chart at the latest saved Pine version. It reuses a current instance, adds a missing one, or transactionally adds the latest version, migrates the analysis inputs, verifies them, then removes the old instance. Source, symbol, timeframe, pine_id, version and input contract are checked fail-closed. Without confirm=true, any chart-changing action is preview-only.

Read the current Bushido Analysis Overlay state without changing the chart. It resolves the study by USER pine_id, verifies the exact on-chart Pine version source and context-bound input contract, then returns analysis metadata, expiry, current-price relations, risk/reward references and drawing integrity. Level states describe only the current price, not historical touch order; use a future outcome tool for that.

Evaluate the first confirmed Target-versus-Stop outcome of an audited Bushido Analysis Overlay using only loaded, closed OHLCV bars after the analysis time. The analysis-containing bar and forming bars are excluded. Entry must precede an optional confirmation; invalidation reached before confirmation cancels the setup. Same-bar ordering, gaps and incomplete history are reported as ambiguous or incomplete rather than guessed. Calendar-month charts are not evaluable because their duration varies. By default it is read-only. When evaluation_timeframe is specified, it temporarily changes only the selected chart's timeframe, captures evidence, and restores the original timeframe; restoration failures are returned explicitly.

Preview or evaluate due analyses directly from the local analysis journal. Without confirm=true it only lists candidates and estimated chart changes. With confirmation it temporarily changes one selected chart's symbol and evidence timeframe for each analysis, evaluates closed OHLCV, records the result, and restores the original chart after every item. Individual failures do not stop the batch; a chart restoration failure stops all remaining work.

Preview or apply a structured market analysis to an existing Bushido Analysis Overlay study. The tool fails closed when the chart symbol, timeframe or overlay input contract does not match. Without confirm=true it is read-only and returns a preview. With confirm=true it changes only that overlay's inputs, then reads them and its drawing totals back for verification. It never places orders or alerts.

Read locally journaled analysis definitions and their monotonic latest evaluations. A completed evaluation is never displaced by a later stale ongoing read. This tool does not access or change the TradingView chart.

Aggregate live analysis-journal outcomes into explicit populations for binary win rate, gross/net R, MFE/MAE and timing. Missing historical path metrics, non-binary outcomes and absent cost assumptions are excluded with counts rather than filled with zero. This read-only tool never mixes Strategy Tester backtests into the live-analysis population.

Calculate confidence calibration from the local analysis journal. Only target_before_stop is labelled positive and stop_before_target negative; ambiguous, incomplete, cancelled, neutral, and unevaluated analyses are reported as exclusions.

Read the backtest report (Strategy Tester) of the strategy currently on the active TradingView chart: net profit, win rate, profit factor, drawdown, Sharpe/Sortino, and the most recent trades with entry/exit details. Percent-style fields are fractions (0.33 = 33%). Fails if no strategy is on the chart — use run_backtest to test a saved strategy without leaving it on the chart.

Record a user-reported research data access in a local append-only journal. Requires confirm:true. Use a stable series_id across revisions and research projects; data_version is a content hash. Records the full inspected UTC interval [from,to), purpose and actual accessed_at. Retries with the same access_id are idempotent; conflicts fail. No chart access, orders or file paths. Reporting use is not preregistration or proof of an unused OOS period.

Compare caller-supplied SHA-256 manifests for previous/current research data, code, runner, candidate rule, parameters and environment. Reports changed and unknown axes plus required revalidation checks. Missing values never match. Does not read files, authenticate hashes, execute code, persist records or certify compatibility, calibration, unused OOS or candidate eligibility.

Check recorded research data usage for a UTC interval [from,to). Finds overlapping use of the same stable series_id across ALL research IDs and data versions. No recorded overlap means unknown outside this journal, never unused or approved OOS. A declared-unused assertion is user supplied and cannot prove unused status. Does not record an access or reserve a period.

Summarize a locally registered immutable backtest ledger by SHA-256 artifact ID. Filter exact symbols, direction, and exit timestamps (UTC, from inclusive/to exclusive); group by symbol/year/month. Requires explicit flat round-trip cost in bps. Recomputes PF from trade-level net wins/losses, never averages PFs. Reports sample break-even flat round-trip cost and signed cost headroom for each population; not an execution guarantee or confidence bound. Returns the entire-ledger baseline, selected subset and excluded complement at the same cost, with complete-case common-opportunity deltas. All filters including dates define selection, not the baseline. Missing outcomes remain missing. No chart access, orders, imports, or arbitrary file paths. Register normalized direction-adjusted gross-bps evidence with the local import CLI first. Optional research_id explicitly enables a local append-only slice exploration journal write before returning metrics; it also records automatic period usage for the entire ledger trade envelope, not underlying price lookbacks. Optional usage_access_id with research_id makes period-record retries idempotent; slice call counts still increment. omitting it is explicitly untracked. Recording failure returns an error without metrics. Counts cover recorded calls for this research ID and artifact only, not all searches or independent trials.

Read a stable, paginated ledger of every trade available in the active Strategy Tester report. Returns entry/exit, direction, profit, quantity, duration, and run-up/drawdown/commission when TradingView exposes them. Start with offset 0; pass the returned ledgerId as expected_ledger_id on later pages to fail closed if the strategy recalculates. Read-only and requires an active strategy.

Run one bounded baseline-versus-candidate Strategy Tester experiment on the active chart. Both variants are resolved to exact saved Pine versions, applied serially with optional input overrides, bound to full-ledger SHA-256 ids, and removed after collection. Returns metric deltas without a synthetic score. Without confirm=true it only previews the experiment. It never places orders.

Run a bounded, serial matrix of saved Pine strategies across explicit symbol, timeframe, and input combinations. The matrix is limited to 24 jobs and a soft runtime budget. Each strategy is temporarily added, bound to a full-ledger SHA-256 id, removed, and the original chart state is restored after every job. Failures and insufficient samples remain as rows; results are never ranked. Without confirm=true this only returns the resolved execution plan. It never places orders.

Run a bounded Pine Strategy walk-forward evaluation from full, immutable trade ledgers. Two to eight exact saved strategy/input candidates are collected serially on the bound chart, then partitioned into two to twelve explicit train, embargo, and test windows by closed-trade entry/exit time. Selection uses train metrics only; only the selected candidate's test metrics are exposed. Candidate failure, ledger quality issues, cost-condition differences, ties, and insufficient samples are not silently ignored. Without confirm=true this only previews the plan. Never places orders.

Run bounded robustness tests for one exact saved Pine strategy. After a dry-run preview, confirm=true collects a baseline full ledger, applies modeled cost, commission, period-start, and seeded bootstrap scenarios, and can serially rerun up to eight explicit Pine input-override scenarios for Entry-delay, Stop/Target, or parameter-neighbor effects. Every temporary Strategy is removed and the chart fingerprint is checked before continuing. Results include failures, distributions, worst cases, and degradation without ranking, adoption, fabricated fills, or orders.

Register one immutable strategy-research hypothesis and its evaluation contract in a local append-only journal. This does not access TradingView or run a test. Reusing a hypothesis_id with a different definition is rejected.

Register one immutable event-study or observational feature-study hypothesis and its outcome contract in the local append-only research journal. Use the optional journal field of run_market_event_study or compute_feature_outcome_relationships to record evidence without copying results by hand.

List append-only market-event and observational feature-study evidence records, or compare exact selected records when both study_ids and evidence_hashes are supplied. It never accesses a chart.

Append one exact strategy experiment result to the research journal. The record binds the hypothesis, population, Pine versions, full-ledger ids, known metrics, guardrails, and decision. It stores no OHLC or source code and never touches a chart.

Compare two to twenty exact saved experiment-evidence records without ranking or combining incompatible populations. References must include both experiment_id and evidence_hash. Read-only and does not access TradingView.

Backtest one of the user's OWN saved strategies on the active chart's current symbol and timeframe: temporarily applies the strategy, waits for the Strategy Tester report, returns it (same shape as get_strategy_report) and removes the strategy again so the chart is left unchanged (set keep_on_chart to keep it). Get strategy pine_ids (kind 'strategy') from list_pine_scripts. Combine with set_symbol/set_timeframe to test other markets.

List the user's TradingView price alerts (read-only): symbol, condition, active state, last fire time. This tool does not create, modify, restart, or delete alerts.

Preview or create bounded, one-shot TradingView price alerts for Confirmation, Invalidation, and Target 1 from one audited Bushido Analysis Overlay. It verifies the exact Pine source, chart binding, analysis_id, current price, existing owned alerts, and post-create readback. Without confirm=true it is read-only. It never uses webhooks, email, SMS, broker APIs, or orders.

Get the user's TradingView watchlists: list names and their symbols, grouped by the user's section headers. Uses the app's logged-in session.

Get current quotes and technical data for specific symbols via TradingView's scanner API (no chart interaction). Default columns include close, change, volume, RSI and 'Recommend.All' — the overall technical rating in [-1, 1] (-1 strong sell, +1 strong buy). Other fields (e.g. 'MACD.macd', 'EMA50', 'price_earnings_ttm') can be requested via columns.

Build a read-only, point-in-time market data snapshot for environment analysis. It fetches quotes and a shared multi-timeframe overview for target and optional auxiliary symbols, with optional economic events. The result explicitly reports missing data, request-time timestamps, and quality status; it does not produce a trade instruction. TradingView scanner values do not expose a common source timestamp, so intraday timing must be treated as partial-quality evidence.

Observe read-only TradingView scanner bid/ask conditions for one or more symbols. It normalizes spread and tick/pip units, rejects crossed quotes and delayed data, and reports ready only when a streaming bid/ask change is observed after the request. It does not access accounts or create, modify, or execute orders.

Build one read-only evidence bundle for trade analysis from a symbol-bound TradingView chart, OHLCV, key levels, scanner MTF/quotes, events, COT, U.S. real yield, and bid/ask execution evidence. decision_status reports only data and gate readiness; it never produces a directional recommendation or changes charts, Pine, alerts, orders, or journals.

Get the latest official U.S. Treasury 10-year par real yield. This is end-of-day macro context, not an intraday trigger. Pass as_of to read only locally persisted versions first seen by that time.

Read only locally first-seen policy-rate versions that were available by the requested time. TradingView decision-date bars are never treated as intraday publication timestamps; this is macro context, not a trading trigger.

Read separately stored official revised policy-rate history for exploratory research only. It never asserts historical availability and must not be used as prospective or out-of-sample carry evidence.

Check whether locally first-seen policy-rate evidence can support a fixed carry panel before any return study runs. It uses only first-seen, available-at policy rates, counts non-overlapping business-day anchors, and returns not_evaluable instead of inventing history.

Measure actual pairwise return correlation and block-bootstrap design effect for a fixed daily FX panel before a carry primary test. It temporarily switches the requested chart for each pair, loads only the requested history, and restores it after every pair. This measures price-panel dependence only; it does not invent policy-rate vintages or run a carry-return study.

Read whether the local OANDA retail-flow collector is configured, without making a network request or exposing credentials. The source is limited to OANDA-client order and position percentages, not market-wide order flow.

Read the frozen carry_core_primary_v1 collection readiness without switching a chart. It combines first-seen policy rates with complete collection heartbeats, rounds the evidence start onto the frozen 20-business-day anchor grid, and reports current heartbeat-gap status.

Run the frozen carry_core_primary_v1 specification on the five-pair core FX panel. It uses only locally first-seen policy-rate versions that were available by each anchor-date close, fits pair fixed effects, and refits the regression in anchor-date moving-block bootstrap samples. It returns not_evaluable until the pre-registered first-seen history has at least 60 complete anchor-date clusters.

Estimate the precision-equivalent sample size of an already fixed carry panel before its primary test. It resamples whole anchor-date cross-sections in circular moving blocks, preserving same-date pair dependence and serial dependence. It is a planning diagnostic only, never an adoption rule or a return-study runner.

Get CME's Daily Bulletin aggregate open interest for COMEX Gold futures (TOTAL GC FUT). This is an exchange-wide all-listed-month total, independent from TradingView chart indicators. The observation is appended to the local first-seen history so a later analysis can distinguish what was initially published from a final revision. It never changes a chart, Pine script, alert, or order.

Combine one exact TradingView CME/COMEX/CBOT/NYMEX continuous-futures daily chart with delayed CFTC COT positioning, trailing volume z-scores, daily Open Interest (OI) 4-quadrant analysis, and contract roll anomaly detection. This is a market-participation proxy, not realtime institutional order flow, and it never changes the chart.

Get the audited Pine overlay that renders a supplied COT crowding context and the chart's prior daily structure range. COT values are explicit MCP inputs; the overlay does not fetch or infer CFTC data, orders, stops, or execution flow.

Describe an exploratory COT crowded-position unwind proxy on a bound daily FX chart. It combines leveraged-money three-year positioning percentiles with a prior-20-day price break; it never claims to observe orders, stops, institutions, or execution flow.

Get the latest or recent public CFTC COT positioning proxy for a supported FX or gold symbol. COT is weekly, delayed futures data, not a realtime order-flow signal.

Reconcile CFTC Gold COT open interest with locally first-seen official CME Daily Bulletin TOTAL GC FUT open interest. Only identical report/observation calendar dates are compared; missing dates are returned as quality evidence and are never nearest-date matched or filled.

Classify rolling return correlation between two explicitly bound layout charts. Closed bars must share an exact UTC timestamp; missing bars are never forward-filled. This is descriptive evidence, not a trading signal.

Coverage preflight for the cross-asset shock study. It binds an EURUSD or USDJPY target chart, then temporarily reuses one auxiliary chart for DXY, US10Y, and XAUUSD at the same 5 or 15 minute timeframe, restoring that auxiliary chart after each read. An optional explicit history load can add older bars but never changes symbols or timeframes persistently. Only exact UTC closed-bar intersections are counted; no shock threshold, direction, outcome, order-flow claim, or trading instruction is produced.

Classify observed EURUSD or USDJPY shocks after exact-time cross-asset collection. It temporarily uses one auxiliary chart for DXY, US10Y, and XAUUSD, restores it after each read, and applies the frozen same-UTC-slot baseline contract. It reports observed state only, never a forward outcome, candidate, order-flow claim, or trading instruction.

Measure descriptive 15/30/60/120-minute outcomes for frozen observed cross-asset shock states. It uses exact-time common bars, excludes overlapping state windows, and never produces a candidate or trade instruction.

Read a local Bookmap Collector session and verify its conservative, receipt-time-only join coverage to the active EURUSD M1 or M5 chart. Bookmap CME flow is retained as a single-venue futures proxy, never presented as spot-FX-wide flow. This preflight creates no candidate, outcome study, chart change, or order.

Measure forward outcomes for caller-supplied point-in-time labels on the active chart, so an external series such as daily open interest, settlement statistics or survey data can be tested with the same outcome, fold, interval and journal machinery as the built-in conditions. Each label is attached to a bar at least one bar after the one it carries, because an external label is rarely public when its own bar closes; a zero lag is refused rather than trusted. Horizons count subsequent observed bars, so a daily series is not voided by weekends. Unlike run_market_event_study this accepts daily and weekly timeframes. Label correctness and revisions belong to whoever supplied them; only the join is point-in-time here. Signal-bar close is an event reference, not an assumed fill. It never ranks labels, changes the chart, or places orders.

Scan return correlation between two explicitly bound layout charts across a symmetric range of lags, so a caller can see whether a reference market leads the primary. Closed bars must share an exact UTC timestamp; missing bars are never forward-filled. The v3 default scales each return by its own series' prior-20-return RMS without using the current return. Every scanned lag is returned with its own interval and per-fold stability, and no best lag is selected. Shared clustered-volatility calibration still exceeds nominal alpha, so candidate eligibility and journal promotion remain disabled. Descriptive evidence only, never a trading signal.

Align closed OHLCV bars already loaded in two or more TradingView layout charts on their exact UTC timestamps. It never changes charts and never forward-fills; forming bars, different resolutions, missing bars, and retrieval failures are reported as quality conditions. Use a layout where each required market is already open.

Validate a frozen strategy-research protocol before adoption decisions. Resolves and statically audits one exact saved Pine strategy version, then checks IS/OOS overlap, future windows, forming-bar use, candidate multiplicity, minimum trades, explicit costs, restart-difference evidence, and definition changes after OOS access. This is read-only and does not run a backtest, inspect the chart, or prove non-repainting.

Statically audit one of the user's Pine scripts for constructs that can repaint or make realtime values differ from historical values. This is a source-level screen, not proof of non-repainting; every result remains restricted until restart-difference validation is recorded by the evaluation pipeline.

Compare two closed-bar captures of the same indicator after a chart reload/restart. Returns changed plot values; it does not persist either capture.

Compute deterministic, non-directional features from exact-time-aligned closed OHLCV observations, typically returned by get_aligned_history: close-to-close return, realized volatility, ATR, and return correlations. It does not fetch data, fill gaps, or produce a trade recommendation.

Compute explicit round-trip spread, slippage, and commission cost for a supported instrument. This is an assumption model, not broker execution data.

Compute a risk-budgeted instrument quantity from entry, stop, explicit execution cost, quantity constraints, and fresh quote-to-account currency evidence. The quantity is always rounded down and the tool fails closed when conversion evidence or minimum-size capacity is missing. It does not access an account or place an order.

Validate a proposed trade plan without changing TradingView, Pine, alerts, orders, or journals. Returns structured quality issues and cost-adjusted Target 1 risk/reward.

Multi-timeframe overview of one or more symbols WITHOUT touching the user's chart: the same indicator fields (default: close, RSI, ADX, ATR, EMA20, SMA50, SMA200 and Recommend.* ratings) across several timeframes in a single call — pass several symbols to compare majors side by side in one call. Use this for top-down analysis (e.g. 1D trend, 240/60 timing) before or instead of set_timeframe.

Screen a market for symbols matching field filters via TradingView's scanner API, e.g. RSI < 30 sorted by volume. Markets: 'america', 'japan', 'crypto', 'forex', 'global', etc. Filter fields use scanner names like 'RSI', 'close', 'volume', 'change', 'market_cap_basic', 'Recommend.All'.

Aggregate key price levels near the current price from ALL price-scale indicators on a TradingView chart: plot values whose titles name a level (S/R, pivot, VWAP, bands, BOS/CHoCH...), horizontal lines, box/zone edges and label prices, each tagged with its source indicator. Oscillator panes (RSI etc.) and generic value plots (open/high/low/close mirrors) are excluded. Sorted by distance from the current price. Use this instead of manually combining get_indicator_values and get_indicator_graphics when you need a support/resistance table.

Get economic calendar events (CPI, NFP, central bank decisions, PMIs...) from TradingView's public calendar: scheduled time, country/currency, importance, forecast/previous/actual values. Use this to check whether upcoming news could invalidate a technical setup. Defaults: next 7 days, medium+ importance, US/EU/JP/GB.

Change one TradingView chart to a different symbol, e.g. 'BTCUSD', 'OANDA:EURUSD', 'NASDAQ:AAPL'. chart_index selects a pane in multi-chart layouts; the active chart is used by default. The target pane is read back and failures are rolled back.

Change one TradingView chart's timeframe. Examples: '1', '5', '15', '60', '240' (minutes), '1D', '1W', '1M'. chart_index selects a pane in multi-chart layouts; the target pane is read back and failures are rolled back.