> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.ninjatrader.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.ninjatrader.com/_mcp/server.

# Skill Reference

**Beta, pre-release.** The NinjaTrader MCP server and its trading skills are in beta and
aren't yet generally available. The beta runs on Demo only, which trades simulated money.
Live access opens later. Server hostnames, the tool set, the skills, and the onboarding
flow may change before launch.

Each skill routes on natural language—you never invoke one by name. The phrases below are examples of what triggers each skill. Anything close routes the same way. Every skill leaves execution decisions to you. See [Trading Skills](/mcp/skills) for details. The Tools column lists the primary [MCP tools](/mcp/tools) each skill uses. Skills also call `user_profile` to discover your accounts when the conversation hasn't named one.

## Pre-trade

| Skill            | What it does                                                                                                                                                                                                      | Ask it things like                                                 | Tools                                                                  |
| ---------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------ | ---------------------------------------------------------------------- |
| `contract-intel` | Resolves bare product codes like "ES" to the tradeable front-month contract by open-interest ranking, classifies rollover status, and knows product families, such as full ES versus micro MES.                   | "Which contract to trade?" · "What's the front month?"             | `search_contracts`, `market_snapshot`                                  |
| `pretrade-risk`  | Sizes a proposed trade from your risk budget and stop distance, then proposes an OCO bracket—stop plus a target at an R-multiple—for the entry. Flags trades that exceed your daily loss budget or contract caps. | "Size a trade on NQ with a 30-point stop." · "How many contracts?" | `my_portfolio`, `market_snapshot`, `risk_settings`, `estimate_order`   |
| `risk-coach`     | Surfaces behavioral context from your own recent fills: revenge-trading patterns, consecutive-loss streaks, overtrading, size drift. Advisory only; it never blocks.                                              | "Is this revenge trading?" · "Sanity check."                       | `fill_history`, `my_portfolio`, `performance_summary`, `risk_settings` |

`contract-intel` resolves the exact symbol the other skills need, so bare product codes route through it first. `pretrade-risk` handles *initial* entry sizing—to change a position you already have, use `scale-manager`. And `risk-coach` answers "is this a good trade?" with behavioral context from your own history, not a yes/no verdict.

## Live position

| Skill               | What it does                                                                                                                                                                                                             | Ask it things like                                               | Tools                                                                                         |
| ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------- | --------------------------------------------------------------------------------------------- |
| `position-watchdog` | Live health check on open positions: open P\&L in dollars and R, distance to stop, progress to target, time in trade, margin utilization, and guardrail flags like a missing stop. Includes what-if math for stop moves. | "Position health." · "Margin usage." · "Daily loss."             | `my_portfolio`, `market_snapshot`, `order_history`, `risk_settings`                           |
| `scale-manager`     | Scales into or out of an existing position—take a partial, pyramid a winner, move the stop to breakeven, trail it, or ladder an entry—and proposes the exact order payloads for you to approve.                          | "Take half off." · "Move stop to breakeven." · "Ladder into ES." | `my_portfolio`, `market_snapshot`, `order_details`, `estimate_order`                          |
| `market-context`    | Computes session VWAP, the value area—POC, VAH, VAL—ATR, realized volatility, delta and order-flow imbalance, and intra-day relative strength, then narrates the market structure.                                       | "What's the VWAP on ES today?" · "What's the volatility regime?" | `market_history`, `market_snapshot`, `search_contracts`                                       |
| `alerts-composer`   | Translates natural-language alert intent into a validated alert expression, checks the syntax offline, and submits it for you.                                                                                           | "Set a trigger." · "Alert."                                      | `create_alert`, `list_alerts`, `my_portfolio`, `market_snapshot`, `search_contracts`          |
| `event-watch`       | Tracks the high-volatility economic calendar—FOMC, CPI, NFP, EIA—filters upcoming events by your positions, and contextualizes moves after they happen.                                                                  | "What's coming up this week?" · "Why did ES just spike?"         | `economic_calendar`, `earnings_calendar`, `my_portfolio`, `market_snapshot`, `market_history` |
| `correlation-hedge` | Computes historical correlation and beta between two futures symbols and turns that into a dollar-neutral hedge proposal, with residual-exposure math. Informational only.                                               | "What's the correlation between ES and NQ?" · "Hedge this."      | `market_history`, `market_snapshot`, `my_portfolio`, `search_contracts`                       |

A bare positions request routes to `position-watchdog`—charting requests go to `chart-render` only when you ask for a chart. `alerts-composer` is also the submission path for alert expressions proposed by `position-watchdog`, `event-watch`, and `pretrade-risk`. For same-underlying pairs like ES and MES, use `contract-intel` rather than `correlation-hedge`. And when you approve a hedge and want it placed, `scale-manager` proposes the payloads.

## Post-trade

| Skill           | What it does                                                                                                                                                                                                                | Ask it things like                                           | Tools                                                                                                                                   |
| --------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------- |
| `trade-journal` | Descriptive review of closed trades: win/loss streaks, hold-time distributions, by-symbol and by-hour P\&L, and per-fill slippage. "What happened today," not "what should change."                                         | "Best and worst trades?" · "Win rate by hour-of-day?"        | `fill_history`, `order_history`, `order_details`, `position_history`, `performance_summary`, `market_history`, `timeline_report`        |
| `trade-debrief` | Educational session review that mines your trading for mistakes, patterns, and correction rules: dollar impact first, every claim cited to the underlying report. Proposed changes are hypotheses, not personalized advice. | "What went wrong?" · "Review today's session."               | `timeline_report`, `performance_summary`, `fill_history`, `position_history`, `market_history`, `economic_calendar`, `search_contracts` |
| `trade-replay`  | Deep dive on one closed trade—reconstructs how far it ran for and against you as MFE and MAE, then replays alternate exits—fixed targets, trailing stops, breakeven pins—against the same bars.                             | "Replay that trade." · "How far did it go before reversing?" | `fill_history`, `position_history`, `market_history`, `search_contracts`                                                                |

The three review skills split by question. `trade-journal` is descriptive: what happened today. `trade-debrief` is prescriptive: what should change. `trade-replay` is single-trade counterfactuals: what if that trade had gone differently. `trade-replay` isn't backtesting—it replays your actual fills, not a strategy over market history.

## Visualization

| Skill          | What it does                                                                                                                                                                                                  | Ask it things like          | Tools                            |
| -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------- | -------------------------------- |
| `chart-render` | Renders candlestick charts with VWAP, fill markers, and price levels; volume-profile histograms with POC and value-area lines; and equity curves with drawdown shading, as PNG images. Requires `matplotlib`. | "Show the chart." · "Plot." | `market_history`, `fill_history` |

`chart-render` draws data the other skills produce: bars and fills directly from the tools, plus analytics from `market-context` and round-trip trades from `trade-journal` when a chart needs them.