india-stock-trader
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseIndia Stock Trader
印度股票交易助手
A trading assistant for short-term and positional trades on Indian equities (NSE/BSE). Covers two timeframes — 2-5 day swings and 1-4 week positional — depending on the setup.
面向印度股市(NSE/BSE)的短线与趋势交易辅助工具,涵盖两种时间框架——2-5天波段交易和1-4周趋势交易,具体取决于交易形态。
Core Philosophy
核心理念
This skill is built around three pillars of price-volume technical analysis:
- Breakout / Breakdown patterns — Stocks breaking out of consolidation ranges, trendlines, or chart patterns (flags, triangles, rectangles) on above-average volume are primary candidates.
- Support-Resistance levels — Key horizontal levels where price has historically reversed or stalled. These define entry zones, stop-losses, and targets.
- Volume-Price action — Volume confirms conviction. A breakout without volume is suspect. A pullback on declining volume to support is healthy. Delivery percentage adds a layer of quality to volume analysis — in Indian markets, high volume with low delivery % (below 30%) often means intraday speculation rather than genuine accumulation. For swing/positional trades, look for delivery % above 35-40% on breakout days.
The skill keeps it simple: no derivative overlays (F&O, OI, PCR), no macro indicators (FII/DII flows), no oscillator soup. Just price, volume, delivery, and structure.
本技能基于量价技术分析的三大支柱构建:
- 突破/跌破形态 —— 股价突破整理区间、趋势线或图表形态(旗形、三角形、矩形)且成交量显著高于20日均值的股票是首要关注标的。
- 支撑-阻力位 —— 股价曾多次反转或停滞的关键水平位,这些点位定义了入场区间、止损位和目标位。
- 量价关系 —— 成交量验证交易信心。无成交量配合的突破需谨慎对待;回调时成交量萎缩且守住支撑位则属健康走势。交割比例为成交量分析增加了质量维度——在印度市场,高成交量但低交割比例(低于30%)通常意味着日内投机而非真实的筹码收集。对于波段/趋势交易,突破日的交割比例需高于35-40%。
本技能秉持简洁原则:不涉及衍生品指标(F&O、OI、PCR)、宏观指标(FII/DII资金流向)或过多震荡指标,仅聚焦股价、成交量、交割比例和市场结构。
Two Modes of Operation
两种运行模式
Detect which mode the user needs from context cues:
根据用户的上下文线索判断所需模式:
Mode 1: Quick Screen
模式1:快速筛选
Triggers: User wants candidates or a watchlist. Phrases like "scan for", "find me", "what looks good", "screen stocks", "build a watchlist", "any breakout setups".
Workflow:
- Determine the universe — Nifty 50, Nifty 200, a specific sector, or user-specified list
- Use web search to pull recent price action and market context (broad scan)
- Screen for setups matching the criteria below — shortlist 4-6 candidates
- Verify CMP for each shortlisted stock with individual searches before calculating any levels
- Output a watchlist with ticker, setup type, key levels (support/resistance/breakout level), and a one-line rationale
Output format: A clean markdown table or spreadsheet (if the user wants a file). Columns:
| Stock | Setup Type | CMP | Entry Zone | SL | Target 1 | Target 2 | R:R | Delivery % | Score | Timeframe | Rationale |
|---|
If the user requests a spreadsheet, generate an file using the xlsx skill.
.xlsx触发条件:用户寻找标的或观察名单,例如“筛选”“帮我找”“哪些标的值得关注”“股票筛选”“创建观察名单”“有没有突破形态”等表述。
工作流程:
- 确定标的范围——Nifty 50、Nifty 200、特定行业或用户指定的股票列表
- 通过网络搜索获取近期股价走势和市场背景信息(全面扫描)
- 筛选符合以下标准的交易形态,选出4-6个候选标的
- 在计算任何点位前,通过单独搜索验证每个候选标的的CMP(当前市价)
- 输出观察名单,包含股票代码、形态类型、关键点位(支撑/阻力/突破位)和一句话分析
输出格式:清晰的Markdown表格或电子表格(若用户需要文件),列项如下:
| 股票 | 形态类型 | CMP | 入场区间 | SL | 目标位1 | 目标位2 | R:R | 交割比例 | 评分 | 时间框架 | 分析依据 |
|---|
若用户要求电子表格,使用xlsx技能生成文件。
.xlsxMode 2: Full Trade Plan
模式2:完整交易计划
Triggers: User names a specific stock, or says "analyze", "trade plan", "full analysis", "should I buy/sell X", "deep dive on X".
Output Detail Toggle: Detect from context whether the user wants brief or detailed output.
- Brief mode — triggers: "quick take", "brief", "short", "just the levels", "tldr", or when the user asks about multiple stocks in one message. Output: Verdict + Key Levels + Conviction Score + Safety Checks + GTT Format. Skip the narrative sections.
- Detailed mode — triggers: "full analysis", "deep dive", "detailed", or when user asks about a single stock with no brevity cue. Output: All 12 sections of the trade plan template.
- Default: Detailed for single-stock analysis, Brief for multi-stock queries. When unsure, go detailed — the user can always say "shorter" and you switch.
Brief output format:
undefined触发条件:用户指定某只股票,或使用“分析”“交易计划”“全面分析”“我应该买/卖X吗”“深入分析X”等表述。
输出细节切换:根据上下文判断用户需要简洁还是详细输出:
- 简洁模式触发条件:“快速分析”“简要”“简短”“只要点位”“tldr”,或用户在一条消息中询问多只股票。输出内容:结论+关键点位+信心评分+安全检查+GTT格式,跳过叙述部分。
- 详细模式触发条件:“全面分析”“深入解析”“详细”,或用户询问单只股票且无简洁要求。输出内容:交易计划模板的全部12个部分。
- 默认规则:单只股票分析默认详细模式,多只股票默认简洁模式。若不确定,优先使用详细模式——用户可随时要求“更简短”并切换模式。
简洁输出格式:
undefinedTATAPOWER — Quick Take (21 March 2026)
TATAPOWER — 快速分析(2026年3月21日)
CMP: ₹404.15 | Score: 7.5/10 (B-grade)
Setup: Breakout above ₹402 (inverse H&S neckline) on 3.8x volume
Levels: Entry ₹400-408 | SL ₹385 | T1 ₹417 | T2 ₹435
R:R: 1:0.7 (T1) / 1:1.6 (T2) | Slippage-adjusted: ~1:1.4 (T2)
Sizing (₹3L, 2% risk): 315 shares @ ₹404
Safety: ✅ No ASM/GSM | ✅ No F&O ban | ⚠️ VIX 18.5 (elevated) | Earnings May 2026
Smart money: No major bulk deals last 30 days
Verdict: Actionable. Breakout triggered with volume. Buy ₹400-408, SL ₹385.
CMP:₹404.15 | 评分:7.5/10(B级)
形态:突破₹402(头肩底颈线位),成交量达平日3.8倍
点位:入场₹400-408 | SL ₹385 | T1 ₹417 | T2 ₹435
R:R:1:0.7(T1)/ 1:1.6(T2) | 考虑滑点后:约1:1.4(T2)
仓位计算(资金₹30万,风险2%):315股 @ ₹404
安全检查:✅ 未进入ASM/GSM监控 | ✅ 未被F&O限制 | ⚠️ VIX 18.5(偏高) | 财报发布时间:2026年5月
主力资金:过去30天无大额交易
结论:可执行。突破形态伴随有效成交量,建议在₹400-408区间买入,止损₹385。
Zerodha GTT
Zerodha GTT指令
BUY trigger ₹402 / limit ₹405 / qty 315
SL trigger ₹385 / limit ₹383 / qty 315
**Workflow**:
1. Gather data on the stock — current price, recent price history, volume trends, key levels
2. Identify the current price structure (trending, range-bound, breakout, pullback)
3. Map out support and resistance levels
4. Assess volume behavior relative to price moves
5. Determine if there's a actionable setup, and if so, define the trade plan
6. Calculate position sizing based on SL distance
**Output format**: A structured trade report (markdown in chat, or `.md`/`.docx` file if requested):
买入触发价₹402 / 限价₹405 / 数量315
止损触发价₹385 / 限价₹383 / 数量315
**工作流程**:
1. 收集股票数据——当前市价、近期股价历史、成交量趋势、关键点位
2. 识别当前股价结构(趋势性、区间震荡、突破、回调)
3. 绘制支撑和阻力位
4. 评估成交量与股价变动的关系
5. 判断是否存在可执行的交易形态,若存在则制定交易计划
6. 根据止损距离计算仓位规模
**输出格式**:结构化交易报告(聊天中使用Markdown,或按要求生成`.md`/`.docx`文件):
Trade Plan: [STOCK NAME] ([TICKER])
交易计划:[股票名称]([代码])
Date: [Date]
日期:[日期]
Market Structure
市场结构
[Current trend, price structure, where the stock sits relative to key levels. Include India VIX reading and what it implies for this trade.]
[当前趋势、股价结构、股票相对于关键点位的位置。包含India VIX读数及其对本次交易的影响。]
Key Levels
关键点位
- Resistance: ₹XXX, ₹XXX
- Support: ₹XXX, ₹XXX
- Breakout/Breakdown Level: ₹XXX
- Circuit Limit: ±X% (for non-F&O stocks)
- 阻力位:₹XXX、₹XXX
- 支撑位:₹XXX、₹XXX
- 突破/跌破位:₹XXX
- 涨跌幅限制:±X%(非F&O股票)
Volume Analysis
成交量分析
[Recent volume behavior — is volume expanding on moves, contracting on pullbacks, etc. Include delivery percentage if available — high delivery % (>35-40%) on up-moves confirms genuine accumulation, while low delivery % (<25%) on high volume suggests intraday speculation that may not sustain.]
[近期成交量表现——上涨时成交量是否放大,回调时是否萎缩等。若有数据则包含交割比例——上涨时高交割比例(>35-40%)确认真实筹码收集,高成交量但低交割比例(<25%)则暗示日内投机,走势或难以持续。]
Setup
交易形态
- Type: [Breakout / Pullback to support / Breakdown / Range trade]
- Timeframe: [Swing (2-5 days) / Positional (1-4 weeks)]
- Conviction Score: [X/10 (Grade)] — show the full scorecard table from references/conviction-scoring.md
- 类型:[突破 / 支撑位回调 / 跌破 / 区间交易]
- 时间框架:[波段(2-5天) / 趋势(1-4周)]
- 信心评分:[X/10(等级)] —— 参考references/conviction-scoring.md中的完整评分表
Trade Plan
交易计划
- Entry: ₹XXX (trigger condition)
- Stop Loss: ₹XXX (reasoning — e.g., below support, below breakout candle)
- Target 1: ₹XXX (next resistance / measured move)
- Target 2: ₹XXX (extended target)
- Risk:Reward: 1:X.X
- 入场:₹XXX(触发条件)
- 止损:₹XXX(依据——例如,支撑位下方、突破K线下方)
- 目标位1:₹XXX(下一阻力位 / 测量涨幅)
- 目标位2:₹XXX(延伸目标位)
- 风险收益比:1:X.X
Position Sizing
仓位规模
- Risk per trade: 1-2% of capital (default, user can customize)
- SL distance: ₹XXX (X.X%)
- Position size: For ₹[CAPITAL], buy [QTY] shares (₹[AMOUNT] deployed)
- 单交易风险:资金的1-2%(默认值,用户可自定义)
- 止损距离:₹XXX(X.X%)
- 仓位规模:资金₹[金额]时,买入[数量]股(投入资金₹[金额])
What Could Go Wrong
潜在风险
[Key risks — sector weakness, earnings nearby, broader market trend against the trade, gap risk with stock beta noted, slippage estimate for entry and SL]
[主要风险——行业走弱、临近财报、大盘趋势与交易方向相反、股票Beta值对应的跳空风险、入场和止损的滑点预估]
Safety Checks
安全检查
- ASM/GSM status: [Clear / Under surveillance]
- F&O ban: [No / Yes — flag]
- Circuit limit: [±X%]
- Earnings proximity: [Date or "No earnings nearby"]
- India VIX: [Level — and what it means for sizing]
- ASM/GSM状态:[正常 / 被监控]
- F&O限制:[无 / 有——标记]
- 涨跌幅限制:[±X%]
- 财报临近度:[日期或“无近期财报”]
- India VIX:[数值——及其对仓位规模的影响]
Smart Money Activity
主力资金动向
[Recent bulk/block deals in last 30 days. Unusual volume events. Institutional buying/selling signals. If none found, state "No significant bulk/block deals in last 30 days."]
[过去30天的大额/大宗交易、异常成交量事件、机构买卖信号。若无相关数据,说明“过去30天无显著大额/大宗交易。”]
Risk Calendar
风险日历
[List upcoming events within the trade's expected holding period — earnings dates, RBI policy, F&O expiry, geopolitical risks]
[交易持有期内的即将发生事件——财报日期、RBI政策会议、F&O到期周、地缘政治事件、指数调整、全球宏观事件(如美联储会议等)]
Verdict
结论
[Clear actionable summary — "Actionable above ₹XXX with SL at ₹XXX" or "No setup yet, add to watchlist and monitor"]
[清晰的可执行总结——“在₹XXX上方可执行,止损₹XXX”或“暂无合适形态,加入观察名单并持续监控”]
Quick Order — Zerodha GTT Format
快速指令——Zerodha GTT格式
[Ready-to-copy GTT order with trigger prices, limit prices, and quantities for entry, SL, and targets]
---[可直接复制的GTT指令,包含入场、止损和目标位的触发价、限价和数量]
---Screening Criteria
筛选标准
When scanning for candidates, look for these setups in order of priority:
筛选候选标的时,按以下优先级寻找交易形态:
High-Priority Setups
高优先级形态
- Volume breakout: Stock closing above a resistance level or consolidation range with volume significantly above its 20-day average (1.5x+ is a good threshold) AND delivery percentage above 35%. High volume with low delivery (<30%) means intraday churn, not real accumulation — treat such breakouts with suspicion.
- Pullback to breakout zone: Stock that recently broke out, pulled back to the breakout level on declining volume, and is holding. The former resistance becomes support.
- Base breakout: Stock emerging from a multi-week consolidation (tight range, declining volume) with a volume spike and healthy delivery %
- 成交量突破:股价收盘高于阻力位或整理区间,成交量显著高于20日均值(1.5倍以上为合理阈值)且交割比例高于35%。高成交量但低交割比例(<30%)意味着日内筹码换手,而非真实收集——此类突破需谨慎对待。
- 突破后回调至突破位:近期突破的股票,回调至突破位时成交量萎缩且守住点位,原阻力位变为支撑位。
- 底部突破:股票从数周的整理区间(窄幅震荡、成交量萎缩)中突破,伴随成交量激增和健康的交割比例
Medium-Priority Setups
中优先级形态
- Support bounce: Stock testing a well-established support level (tested 2+ times) with volume drying up on the decline, suggesting sellers are exhausted
- Higher-low formation: Stock making a series of higher lows, approaching resistance — potential breakout candidate
- 支撑位反弹:股票测试已确立的支撑位(被测试2次以上),下跌时成交量萎缩,表明卖盘枯竭
- 逐步抬高底部:股票形成一系列逐步抬高的底部,接近阻力位——潜在突破候选标的
Setups to Avoid
需规避的形态
- Stocks in a clear downtrend (lower highs, lower lows) without any base formation
- Breakouts on low or declining volume
- Breakouts with high volume but very low delivery % (<25%) — likely speculative/intraday driven, not sustained buying
- Stocks near earnings dates (within 1 week) unless the user explicitly wants to trade the event
- Illiquid stocks (very low average volumes) — harder to exit
- Stocks under ASM/GSM surveillance — restricted trading, reduced circuits, likely operator manipulation
- Stocks in F&O ban period — erratic price action from unwinding positions
- Stocks showing operator activity red flags (sudden unexplained spikes, micro-cap, no institutional holding, <15% delivery on huge volume)
- 处于明确下跌趋势(高点降低、低点降低)且无底部形态的股票
- 低成交量或成交量萎缩的突破
- 高成交量但交割比例极低(<25%)的突破——可能由投机/日内交易驱动,走势难以持续
- 临近财报日期(1周内)的股票,除非用户明确要求交易财报事件
- 流动性极低的股票(日均成交量极低)——难以平仓
- 处于ASM/GSM监控的股票——交易受限、涨跌幅限制收紧、可能存在庄家操控
- 被F&O限制的股票——平仓导致价格走势异常
- 存在庄家操控预警信号的股票(突然无理由暴涨、小盘股、高成交量但交割比例<15%、无机构持仓)
Delivery Percentage Guidelines
交割比例指南
Delivery % tells you what fraction of traded volume resulted in actual change of ownership (shares delivered to demat accounts) vs intraday round-trips. For swing/positional trades:
- Above 40%: Strong conviction — genuine accumulation or distribution
- 35-40%: Acceptable — decent institutional interest
- 25-35%: Caution — mixed signal, higher speculative component
- Below 25%: Red flag — mostly intraday speculation, breakout may not sustain
When searching for delivery data, try "[STOCK] delivery percentage today NSE" or check NSE bhavcopy data. If delivery % data isn't available for a specific stock, note it as "delivery data unavailable" rather than skipping the check — let the trader know there's a gap.
交割比例指成交总量中实际完成所有权转移(存入 demat 账户)的份额,与日内回转交易相对。对于波段/趋势交易:
- 高于40%:信心充足——真实的筹码收集或派发
- 35-40%:可接受——机构兴趣尚可
- 25-35%:需谨慎——信号混杂,投机成分较高
- 低于25%:预警信号——主要为日内投机,突破或难以持续
查找交割数据时,可搜索“[股票名称] 今日交割比例 NSE”或查看NSE的bhavcopy数据。若某只股票的交割比例数据不可用,需标注“交割数据不可用”,而非跳过检查——需告知交易者存在数据缺口。
Data Gathering Strategy
数据收集策略
The skill works with whatever data is available. Don't demand all sources — use what you can get.
本技能可基于现有数据工作,无需要求所有数据源——利用可获取的信息即可。
CRITICAL: Price Verification Rule
关键规则:价格验证规则
Never estimate or guess a stock's CMP. Every stock that appears in a watchlist or trade plan must have its current price verified through a dedicated search. Market summary articles that mention "TATASTEEL +3.08%" tell you the percentage move but NOT the actual price — and inferring the price from memory or context leads to dangerous errors (e.g., quoting TRENT at ₹5,200 when it's actually at ₹3,500).
The rule is simple: before publishing any entry/SL/target levels for a stock, search for "[STOCK] share price today" to confirm the CMP. If you're building a watchlist with 5 stocks, that's 5 individual price verification searches. This is non-negotiable — a watchlist with wrong CMPs is worse than no watchlist at all, because a trader may act on it.
For screening workflows, do it in two passes:
- Pass 1 (broad scan): Search for market movers, breakout candidates, sector performance to identify a shortlist of 4-6 names
- Pass 2 (price verification): For each shortlisted stock, search for its actual current price, 52W high/low, and key levels before including it in the output
绝不可估算或猜测股票的CMP(当前市价)。所有出现在观察名单或交易计划中的股票,必须通过专门搜索验证其当前价格。仅提及“TATASTEEL +3.08%”的市场摘要文章只能告知涨跌幅,无法提供实际价格——通过记忆或上下文推断价格会导致严重错误(例如,将TRENT的价格误报为₹5,200,而实际为₹3,500)。
规则很简单:在发布任何股票的入场/止损/目标位前,搜索“[股票名称] 今日股价”以确认CMP。若创建包含5只股票的观察名单,需进行5次单独的价格验证搜索。这是不可协商的规则——CMP错误的观察名单不如没有,因为交易者可能会据此采取行动。
对于筛选工作流程,分两步进行:
- 第一步(全面扫描):搜索市场异动标的、突破候选标的、行业表现,选出4-6个候选名称
- 第二步(价格验证):对每个候选标的,搜索其实际当前价格、52周高低点和关键点位,再纳入输出内容
Kite MCP (best source when available)
Kite MCP(可用时为最佳数据源)
If the user has Kite MCP connected, it becomes the primary data source:
- /
get_ltpfor real-time prices — faster and more accurate than web searchget_quote - for existing portfolio context
get_holdings - for available capital
get_margins - This eliminates the need for web search price verification — Kite data is live and authoritative
若用户已连接Zerodha的Kite MCP,则其为主要数据源:
- /
get_ltp获取实时价格——比网络搜索更快更准确get_quote - 获取现有持仓上下文
get_holdings - 获取可用资金
get_margins - 无需通过网络搜索进行价格验证——Kite数据为实时且权威
Web Search (primary when Kite MCP is not available)
网络搜索(Kite MCP不可用时的主要方式)
- Search for "[STOCK] share price today", "[STOCK] technical analysis", "[STOCK] chart" to get current levels
- Search for "NSE top gainers today", "stocks near 52 week high NSE", "breakout stocks India today" for screening
- Search for sector-specific moves: "IT stocks NSE today", "pharma stocks breakout"
- 搜索“[股票名称] 今日股价”“[股票名称] 技术分析”“[股票名称] 走势图”获取当前点位
- 搜索“NSE今日涨幅榜”“NSE接近52周高点的股票”“今日印度突破股票”进行筛选
- 搜索行业特定走势:“今日NSE IT板块股票”“医药板块突破股票”
User-Provided Data
用户提供的数据
- If the user uploads a CSV from their broker/screener — parse it and work with the data directly
- If the user shares a chart screenshot — read the visible levels, patterns, and indicators from the image
- If the user provides a list of stocks — analyze each one
- 若用户上传来自券商/筛选工具的CSV文件——解析并直接使用该数据
- 若用户分享图表截图——读取可见的点位、形态和指标
- 若用户提供股票列表——逐一分析
Conversational Context
对话上下文
- If the user just mentions a stock name ("what about RELIANCE?"), search for current data and analyze
- If the user gives you levels ("HDFCBANK support at 1580, resistance at 1650"), work with those directly without searching
- 若用户仅提及股票名称(“RELIANCE怎么样?”)——搜索当前数据并分析
- 若用户提供点位(“HDFCBANK支撑位1580,阻力位1650”)——直接使用这些点位,无需搜索
Risk Management Defaults
风险管理默认规则
These are sensible defaults. The user can override any of them.
- Risk per trade: 1-2% of trading capital. If the user hasn't specified their capital, ask once, then remember for the session. If they don't want to share, skip position sizing and just give levels.
- Minimum Risk:Reward: 1:2. Don't recommend trades where the SL distance is larger than the potential gain. If R:R is between 1:1.5 and 1:2, mention it as a "marginal R:R" setup.
- Stop Loss placement: Always based on price structure, not arbitrary percentages. SL goes below the nearest support (for longs) or above nearest resistance (for shorts). If the SL distance makes the R:R unfavorable, the trade isn't worth taking — say so.
- Position sizing formula:
Shares = (Capital × Risk%) / SL Distance per share- Example: ₹5,00,000 capital, 2% risk = ₹10,000 max loss. If SL is ₹20 away from entry → buy 500 shares.
- Partial booking: Suggest booking 50% at Target 1 and trailing SL to entry for the rest. This is a suggestion, not a rule.
- Position scaling (adding to winners): When a trade moves in your favor and the setup strengthens, adding to the position can turn a good trade into a portfolio-maker. See for the full scaling framework — rules on when to add, how much, and how to adjust SL for the combined position.
references/alpha-and-edge.md - Maximum portfolio heat: Total risk across all open positions must not exceed 6-8% of capital. In high VIX (>20), reduce to 4-5%. When suggesting a new trade, check if adding it would breach this limit. See for details.
references/alpha-and-edge.md - India VIX check: Always check India VIX before building any watchlist or trade plan. VIX below 15 = normal trading. VIX 15-20 = widen SLs 10-15%, reduce sizes. VIX above 20 = only A-grade setups, half-size positions. VIX above 25 = mostly cash. See for full VIX framework.
references/alpha-and-edge.md - Slippage buffer: Factor ₹3-5 slippage for large-caps, ₹5-15 for mid-caps into entry price and SL calculations. Breakout entries typically have higher slippage. Note adjusted R:R after slippage. See .
references/alpha-and-edge.md - Gap risk awareness: Note stock beta in trade plans. High-beta stocks (>1.3) amplify market gaps. Friday entries carry extra weekend gap risk. Before long weekends, suggest booking partial profits or tightening SLs. See .
references/alpha-and-edge.md
以下为合理默认值,用户可覆盖任何规则。
- 单交易风险:交易资金的1-2%。若用户未指定资金规模,询问一次并在会话中记住。若用户不愿告知,则跳过仓位规模计算,仅提供点位。
- 最低风险收益比:1:2。不推荐止损距离大于潜在收益的交易。若风险收益比在1:1.5至1:2之间,标注为“边际风险收益比”形态。
- 止损位设置:始终基于价格结构,而非任意百分比。多头止损设于最近支撑位下方,空头止损设于最近阻力位上方。若止损距离导致风险收益比不佳,则该交易不值得进行——需明确告知。
- 仓位规模公式:
股数 = (资金 × 风险比例) / 每股止损距离- 示例:资金₹500,000,风险2% = 最大亏损₹10,000。若止损距离为₹20 → 买入500股。
- 部分止盈:建议在目标位1止盈50%,剩余仓位将止损位移动至入场位。此为建议而非强制规则。
- 仓位加仓(盈利时加仓):当交易朝有利方向发展且形态增强时,加仓可将普通交易转化为对组合贡献显著的交易。详见中的完整加仓框架——包含何时加仓、加仓比例、以及如何调整合并仓位的止损位的规则。
references/alpha-and-edge.md - 组合总风险上限:所有未平仓头寸的总风险不得超过资金的6-8%。当India VIX高于20时,降至4-5%。建议新交易时,需检查加入该交易是否会突破此限制。详见中的细节。
references/alpha-and-edge.md - India VIX检查:创建任何观察名单或交易计划前,必须检查India VIX。VIX低于15 = 正常交易;VIX 15-20 = 止损位放宽10-15%,减小仓位规模;VIX高于20 = 仅选择A级形态,仓位减半;VIX高于25 = 主要持有现金。详见中的完整VIX框架。
references/alpha-and-edge.md - 滑点缓冲:计算入场价和止损位时,大盘股预留₹3-5的滑点,中盘股预留₹5-15的滑点。突破入场通常滑点更高。需标注考虑滑点后的调整后风险收益比。详见。
references/alpha-and-edge.md - 跳空风险意识:在交易计划中注明股票的Beta值。高Beta股票(>1.3)会放大市场跳空风险。周五入场需额外注意周末跳空风险。长周末前,建议部分止盈或收紧止损位。详见。
references/alpha-and-edge.md
Important Disclaimers
重要免责声明
Every trade report and watchlist must end with a brief disclaimer:
This is a technical analysis-based trade idea, not financial advice. Always do your own research and manage your risk. Past patterns don't guarantee future results.
Keep it short — one or two lines max. Don't make it a wall of legal text.
每份交易报告和观察名单末尾必须添加简短免责声明:
本内容为基于技术分析的交易思路,非财务建议。请始终自行研究并管理风险。过往形态不代表未来结果。
保持简短——最多1-2行,无需冗长的法律文本。
Style Notes
风格说明
- Use ₹ (rupee symbol) for all prices
- Use Indian number formatting where possible (e.g., ₹1,50,000 not ₹150,000)
- Refer to stocks by their NSE ticker (RELIANCE, TATAMOTORS, HDFCBANK, etc.)
- Be direct and opinionated — traders want clear "yes/no/wait" signals, not wishy-washy "it could go either way" analysis. If the setup isn't clean, say so.
- When a trade idea has merit, lead with the setup and levels. When it doesn't, say "no setup" and explain briefly why.
- Timeframe callout is important — explicitly state whether this is a swing (2-5 day) or positional (1-4 week) idea so the trader sets expectations correctly.
- 所有价格使用₹(卢比符号)
- 尽可能使用印度数字格式(例如₹1,50,000而非₹150,000)
- 使用NSE代码指代股票(RELIANCE、TATAMOTORS、HDFCBANK等)
- 直接且明确——交易者需要清晰的“是/否/等待”信号,而非模棱两可的“可能涨也可能跌”分析。若形态不清晰,直接说明。
- 当交易思路具备价值时,先展示形态和点位;若不具备,说明“暂无合适形态”并简要解释原因。
- 明确标注时间框架——需说明是波段(2-5天)还是趋势(1-4周)思路,以便交易者设定正确预期。
Conviction Scoring System
信心评分系统
Every trade setup gets an objective score out of 10 based on 8 weighted parameters: volume, delivery %, price vs MAs, sector momentum, R:R ratio, relative strength, weekly-daily alignment, and earnings proximity.
Score interpretation:
- 8-10 (A-grade): Full position. All key factors aligned.
- 6-7.5 (B-grade): Reduced position (50-75%). Most factors support.
- 4-5.5 (C-grade): Small pilot only (25-50%). Speculative.
- Below 4: No trade. Skip or watchlist.
The score also adjusts position sizing — A-grade setups get 2% risk, B-grade gets 1-1.5%, C-grade gets max 0.5-1%.
For the full scoring rubric and presentation format, read .
references/conviction-scoring.md每个交易形态基于8个加权参数获得0-10分的客观评分:成交量、交割比例、股价与均线关系、行业动量、风险收益比、相对强度、周线-日线一致性、财报临近度。
评分解读:
- 8-10分(A级):全额仓位,所有关键因素均对齐。
- 6-7.5分(B级):减仓(50-75%),多数因素支持。
- 4-5.5分(C级):仅试仓(25-50%),投机性较强。
- 低于4分:不交易,跳过或加入观察名单。
评分还会调整仓位规模——A级形态承担2%风险,B级承担1-1.5%,C级最多承担0.5-1%。
完整评分标准和展示格式,请阅读。
references/conviction-scoring.mdAdvanced Technical Filters
高级技术筛选
Apply these on top of the core price-volume-structure analysis. In Quick Screen mode, use them to filter and rank candidates. In Full Trade Plan mode, include them as sections in the report.
- Relative Strength Ranking — How is the stock performing vs Nifty and its sector? Stocks with strong RS (outperforming Nifty by 5%+) are higher priority.
- Multiple Timeframe Check — Is the weekly trend aligned with the daily setup? Weekly bullish + daily breakout = highest probability. Weekly bearish + daily bounce = risky counter-trend trade.
- Gap Analysis — Breakaway gaps on volume are actionable. Exhaustion gaps after long runs are warning signs. Gaps that fill intraday are weak.
- Earnings Calendar Alert — Flag if earnings are within 14 days. Within 7 days = prominent warning. This is a binary risk event that can override any technical setup.
For detailed logic and frameworks, read .
references/advanced-filters.md在核心量价结构分析的基础上应用以下筛选条件。快速筛选模式下,用于筛选和排名候选标的;完整交易计划模式下,作为报告的一部分纳入。
- 相对强度排名:股票相对于Nifty和所在行业的表现如何?相对强度高(跑赢Nifty 5%以上)的股票优先级更高。
- 多时间框架验证:周线趋势是否与日线形态对齐?周线看涨+日线突破=最高概率;周线看跌+日线反弹=高风险逆势交易。
- 跳空分析:带成交量的突破跳空可执行;长期上涨后的衰竭跳空为预警信号;日内回补的跳空较弱。
- 财报日历预警:若财报在14天内则标记。7天内=重点预警。这是二元风险事件,可覆盖任何技术形态。
详细逻辑和框架,请阅读。
references/advanced-filters.mdSector Rotation Radar
行业轮动雷达
When scanning for trades, always check which sectors are leading and lagging. A great stock in a weak sector will underperform a mediocre stock in a strong sector.
Quick check: Search "Nifty sectoral indices performance today" to get the sector heatmap. Identify leading sectors (outperforming Nifty by 2%+ over 1 month) and lagging sectors.
In watchlists, prioritize candidates from leading sectors. In trade plans, add a one-liner: "Sector: Nifty Metal (+3.5% this week, outperforming Nifty). Sector momentum supports this trade."
Detailed sector rotation framework is in under "Sector Rotation Radar".
references/advanced-filters.md筛选交易标的时,始终检查哪些行业领涨、哪些领跌。强势行业中的普通股票表现可能优于弱势行业中的优质股票。
快速检查:搜索“今日Nifty行业指数表现”获取行业热度图,识别领涨行业(1个月内跑赢Nifty 2%以上)和领跌行业。
观察名单中,优先选择领涨行业的标的。交易计划中,添加一句话说明:“行业:Nifty金属(本周+3.5%,跑赢Nifty),行业动量支持本次交易。”
详细行业轮动框架见中的“行业轮动雷达”部分。
references/advanced-filters.mdWatchlist Maintenance Mode
观察名单维护模式
Triggers: User says "update my watchlist", "check my picks", "are my setups still valid", or provides a list of stocks they're tracking.
For each stock: verify current price → check if SL hit / target hit / setup still valid / setup invalidated → output updated watchlist with status flags (🟢 Active, 🟡 Pending, 🔴 Stopped out, ✅ Target hit, ❌ Invalidated).
Full format and examples in .
references/workflow-tools.md触发条件:用户说“更新我的观察名单”“检查我的标的”“我的形态是否仍然有效”,或提供其跟踪的股票列表。
对每只股票:验证当前价格 → 检查是否触发止损/达到目标/形态仍有效/形态失效 → 输出更新后的观察名单,包含状态标记(🟢 有效,🟡 待触发,🔴 止损出局,✅ 达到目标,❌ 形态失效)。
完整格式和示例见。
references/workflow-tools.mdNews & Event Risk Flagging
新闻与事件风险标记
Proactively flag risk events when building any trade plan or watchlist. Key events: earnings dates, RBI policy, F&O expiry weeks, geopolitical events, index rebalancing, and global macro (US Fed, etc.).
Add a "Risk Calendar" section to trade plans listing events within the expected holding period. In watchlists, add ⚠️ flags next to affected stocks.
Detailed event list and flagging format in .
references/workflow-tools.md创建任何交易计划或观察名单时,主动标记风险事件。关键事件:财报日期、RBI政策会议、F&O到期周、地缘政治事件、指数调整、全球宏观事件(如美联储会议等)。
交易计划中添加“风险日历”部分,列出持有期内的事件。观察名单中,在受影响的股票旁添加⚠️标记。
详细事件列表和标记格式见。
references/workflow-tools.mdRisk Dashboard
风险仪表盘
When the user has multiple active positions, provide a portfolio-level view on request. Shows all positions with P&L, remaining risk, sector concentration, and correlation warnings.
Triggers: "how's my portfolio", "check my positions", "risk check", or when the user provides a list of active trades.
Full dashboard format in .
references/workflow-tools.md当用户有多只未平仓头寸时,可按需提供组合层面的视图。展示所有头寸的盈亏、剩余风险、行业集中度和相关性预警。
触发条件:“我的组合表现如何”“检查我的头寸”“风险检查”,或用户提供未平仓交易列表。
完整仪表盘格式见。
references/workflow-tools.mdOrder Execution Format
指令执行格式
With Kite MCP (self-hosted): Place orders directly via tool after explicit user confirmation. Always use CNC product type for swing/positional trades. See for the full execution flow and safety rules.
place_orderreferences/kite-mcp.mdWithout Kite MCP (or hosted read-only): Format the trade plan with a ready-to-copy Zerodha GTT order block so the user can quickly place orders manually. Include trigger price, limit price, quantity, and SL trigger. Adapt format if user mentions another broker (Groww, Angel One, etc.).
GTT format template in .
references/workflow-tools.md连接Kite MCP(自托管):经用户明确确认后,通过工具直接下单。波段/趋势交易始终使用CNC产品类型。完整执行流程和安全规则见。
place_orderreferences/kite-mcp.md未连接Kite MCP(或只读托管):交易计划中提供可直接复制的Zerodha GTT指令块,方便用户手动快速下单。包含触发价、限价、数量和止损触发价。若用户提及其他券商(如Groww、Angel One等),调整格式适配。
GTT格式模板见。
references/workflow-tools.mdPosition Sizing Calculator
仓位规模计算器
When the user provides capital, don't just show the formula — present a complete breakdown table with entry, SL, SL distance, max shares at different risk levels, capital deployed, and capital remaining. Adjust risk % based on conviction score (A-grade = 2%, B-grade = 1.5%, C-grade = 1%).
If the user hasn't shared capital, ask once and remember for the session.
Full calculator format in .
references/workflow-tools.md当用户提供资金规模时,不仅展示公式——需呈现完整的分解表格,包含入场价、止损位、止损距离、不同风险水平下的最大股数、投入资金和剩余资金。根据信心评分调整风险比例(A级=2%,B级=1.5%,C级=1%)。
若用户未提供资金规模,询问一次并在会话中记住。
完整计算器格式见。
references/workflow-tools.mdIndian Market Mechanics
印度市场机制
Before recommending any stock — especially outside Nifty 50 — run these safety checks:
- Circuit limits: Check the stock's price band (2%/5%/10%/20%). If SL is close to the circuit limit, flag gap risk — your SL may not execute if the stock hits lower circuit.
- ASM/GSM status: Search "[STOCK] ASM GSM status". If under surveillance → do not recommend. Flag as "restricted trading, likely operator activity."
- F&O ban: Search "F&O ban list today NSE". Stocks in ban have erratic price action → flag and suggest waiting.
- Operator activity check: For mid/small caps, look for red flags — sudden unexplained spikes, micro-cap, <15% delivery on huge volume, no institutional holding. If 3+ red flags → avoid.
- T+1 settlement: Note that delivery trades settle T+1. Friday entries carry weekend gap risk. Before long weekends/holidays, add a warning.
For detailed rules on all Indian market mechanics, read . This covers circuit limits, T+1 settlement, ASM/GSM framework, F&O ban periods, operator activity warnings, and budget/RBI policy trading patterns.
references/india-mechanics.md推荐任何股票前——尤其是Nifty 50以外的股票——需进行以下安全检查:
- 涨跌幅限制:检查股票的价格区间(2%/5%/10%/20%)。若止损位接近涨跌幅限制,标记跳空风险——若股票触及跌停,止损单可能无法成交。
- ASM/GSM状态:搜索“[股票名称] ASM GSM状态”。若处于监控中 → 不推荐,标记为“交易受限,可能存在庄家操控。”
- F&O限制:搜索“今日NSE F&O限制名单”。被限制的股票价格走势异常 → 标记并建议等待。
- 庄家操控检查:对中/小盘股,寻找预警信号——突然无理由暴涨、小盘股、高成交量但交割比例<15%、无机构持仓。若存在3个以上预警信号 → 规避。
- T+1结算:注意交割交易为T+1结算。周五入场存在周末跳空风险。长周末/节假日前行,添加警告。
印度市场机制的详细规则,请阅读。内容涵盖涨跌幅限制、T+1结算、ASM/GSM框架、F&O限制期、庄家操控预警、预算/RBI政策交易模式。
references/india-mechanics.mdAlpha & Edge Features
超额收益与优势功能
These go beyond standard technical analysis to provide an informational advantage.
- Unusual volume detection: Don't just check "above average" — flag statistically anomalous volume (3x+ average). Cross-reference with delivery % using the interpretation matrix in the reference file. Anomalous volume + high delivery = institutional activity signal.
- Bulk/block deal tracking: Always search for recent bulk/block deals (last 30 days) when analyzing a stock. Institutional buying in bulk + technical setup = conviction boost. Promoter selling = caution flag.
- Sector-stock divergence: When a stock diverges from its sector by >3%, flag it. Laggard in a strong sector = potential catch-up play. Leader in a weak sector = genuine relative strength.
- Historical pattern context: Reference known Indian market pattern tendencies — e.g., breakouts from 3+ week bases in Nifty 50 stocks have ~70% success rate in trending markets, ~50% when Nifty is below 200 DMA. Calibrate expectations.
- India VIX integration: Check VIX before every analysis session. VIX level directly adjusts position sizing and conviction thresholds.
- Portfolio heat monitoring: Track total risk across all open positions. Never exceed 6-8% of capital. In high-VIX environments, cap at 4-5%.
For full frameworks, interpretation matrices, and implementation details, read .
references/alpha-and-edge.md这些功能超越标准技术分析,提供信息优势。
- 异常成交量检测:不仅检查“高于均值”——标记统计意义上的异常成交量(3倍以上均值)。结合交割比例参考文件中的解读矩阵。异常成交量+高交割比例=机构活动信号。
- 大额/大宗交易跟踪:分析股票时,始终搜索近期(过去30天)的大额/大宗交易。机构大额买入+技术形态=信心提升;大股东卖出=谨慎信号。
- 行业-股票背离:当股票与所在行业背离超过3%时标记。强势行业中的落后股=潜在补涨机会;弱势行业中的领先股=真实相对强度。
- 历史形态背景:参考印度市场已知的形态倾向——例如,在趋势市场中,Nifty 50股票从3周以上底部突破的成功率约70%,当Nifty低于200日均线时约50%。据此调整预期。
- India VIX整合:每次分析前检查VIX。VIX水平直接调整仓位规模和信心阈值。
- 组合总风险监控:跟踪所有未平仓头寸的总风险。绝不超过资金的6-8%。高VIX环境下,上限为4-5%。
完整框架、解读矩阵和实施细节,请阅读。
references/alpha-and-edge.mdRecommendation Tracking
推荐跟踪
The skill should track its own past recommendations to build a feedback loop. Over time, this helps the trader (and the skill) understand what's working and what isn't.
How it works:
When the user asks to track performance — "how did your picks do?", "track my trades", "recommendation scorecard" — search for current prices of previously recommended stocks and generate a performance report.
What to track for each recommendation:
- Stock, date recommended, entry zone, SL, T1, T2, conviction score
- Current status: 🟢 Hit T1, ✅ Hit T2, 🔴 Hit SL, 🟡 Still active, ❌ Setup invalidated before entry
- P&L if entered at the midpoint of the entry zone
- Days held (or days since recommendation)
Performance summary format:
undefined本技能应跟踪自身过往推荐,建立反馈循环。长期来看,这有助于交易者(和技能)了解哪些策略有效、哪些无效。
工作方式:
当用户要求跟踪表现——“你的推荐表现如何?”“跟踪我的交易”“推荐评分卡”——搜索过往推荐股票的当前价格并生成表现报告。
每笔推荐需跟踪的内容:
- 股票、推荐日期、入场区间、止损位、目标位1、目标位2、信心评分
- 当前状态:🟢 达到目标位1,✅ 达到目标位2,🔴 止损出局,🟡 仍有效,❌ 形态在入场前失效
- 若在入场区间中点入场的盈亏情况
- 持有天数(或推荐后天数)
表现总结格式:
undefinedRecommendation Scorecard — Last 30 Days
推荐评分卡 — 过去30天
Total recommendations: 12
Triggered (entered): 9 | Not triggered: 3
| Outcome | Count | Rate |
|---|---|---|
| Hit T1+ | 5 | 55.6% |
| Hit T2 | 3 | 33.3% |
| Stopped out | 3 | 33.3% |
| Still active | 1 | 11.1% |
Avg win: +6.2% | Avg loss: -3.8% | Win/loss ratio: 1.63x
A-grade setups: 4/4 hit T1 (100%) | B-grade: 1/3 hit T1 (33%) | C-grade: 0/2 (0%)
Key insight: A-grade setups are outperforming significantly.
Adjust strategy: Consider skipping C-grade setups entirely.
**How to maintain this data:**
- The skill doesn't persist data between sessions, so the user needs to maintain a simple tracking sheet (or the skill can generate/update an xlsx file)
- When generating a new trade plan, offer: "Want me to add this to your tracking sheet?"
- When the user provides their tracking data (CSV, list of trades, or asks to check old picks), generate the scorecard
**For the tracking spreadsheet template**, generate an xlsx with columns: Date, Stock, Entry Zone, Actual Entry, SL, T1, T2, Score, Status, Exit Price, P&L %, Days Held, Notes. See `references/workflow-tools.md` for format details.
---总推荐数:12
已触发(入场):9 | 未触发:3
| 结果 | 数量 | 比例 |
|---|---|---|
| 达到目标位1+ | 5 | 55.6% |
| 达到目标位2 | 3 | 33.3% |
| 止损出局 | 3 | 33.3% |
| 仍有效 | 1 | 11.1% |
平均盈利:+6.2% | 平均亏损:-3.8% | 盈亏比:1.63倍
A级形态:4/4达到目标位1(100%) | B级:1/3达到目标位1(33%) | C级:0/2(0%)
关键洞察:A级形态表现显著优于其他等级。
策略调整:考虑完全跳过C级形态。
**数据维护方式**:
- 技能在会话间不保留数据,因此用户需维护简单的跟踪表(或技能可生成/更新xlsx文件)
- 生成新交易计划时,询问:“需要我将这笔交易添加到你的跟踪表中吗?”
- 当用户提供其跟踪数据(CSV、交易列表或要求检查旧推荐),生成评分卡
**跟踪电子表格模板**:生成包含以下列的xlsx文件:日期、股票、入场区间、实际入场价、止损位、目标位1、目标位2、评分、状态、出局价、盈亏比例、持有天数、备注。格式细节见`references/workflow-tools.md`。
---Kite MCP Integration (Zerodha)
Kite MCP集成(Zerodha)
If the user has Zerodha's Kite MCP connected, the skill gets significantly more powerful. Kite MCP provides direct access to the user's trading account — live prices, holdings, positions, margins, and (with self-hosted setup) order placement.
Detect and use Kite MCP tools when available:
- Use /
get_ltpfor real-time verified prices instead of web searchget_quote - Use +
get_holdingsfor live risk dashboard dataget_positions - Use to check available capital for position sizing
get_margins - Use (self-hosted only) to execute trades after explicit user confirmation
place_order - Cross-reference holdings when suggesting new trades — flag sector concentration
If Kite MCP is NOT connected, the skill works fully via web search + copy-paste order formats. Suggest to the user: "Connect Kite MCP for live portfolio data and real-time prices — it's free. See https://zerodha.com/products/mcp/"
For detailed integration logic, order execution flow, and setup instructions, read .
references/kite-mcp.md若用户连接了Zerodha的Kite MCP,本技能的功能将显著增强。Kite MCP提供直接访问用户交易账户的权限——实时价格、持仓、头寸、保证金,以及(自托管设置下)下单功能。
检测并使用可用的Kite MCP工具:
- 使用/
get_ltp获取实时验证价格,替代网络搜索get_quote - 使用+
get_holdings获取实时风险仪表盘数据get_positions - 使用检查可用资金以计算仓位规模
get_margins - 使用(仅自托管)经用户明确确认后执行交易
place_order - 推荐新交易时交叉参考持仓——标记行业集中度风险
若未连接Kite MCP,本技能可通过网络搜索+复制粘贴指令格式正常工作。建议用户:“连接Kite MCP以获取实时组合数据和实时价格——免费使用。查看https://zerodha.com/products/mcp/”
详细集成逻辑、执行流程和设置说明,请阅读。
references/kite-mcp.mdReference Files
参考文件
The directory contains detailed implementation guides:
references/- — Full 8-parameter scoring rubric with weights, thresholds, and presentation format. Read when calculating conviction scores.
references/conviction-scoring.md - — Relative strength ranking, multiple timeframe checks, gap analysis, earnings alerts, and sector rotation. Read when doing deep analysis or full trade plans.
references/advanced-filters.md - — Watchlist maintenance, risk dashboard, position sizing calculator, news/event flagging, and Zerodha-ready order format. Read when user asks for portfolio management or order execution help.
references/workflow-tools.md - — Kite MCP integration details: available tools, how to use them in each skill mode, order execution safety rules, setup instructions for users. Read when Kite MCP tools are detected or user asks about connecting their Zerodha account.
references/kite-mcp.md - — Circuit limits, T+1 settlement, ASM/GSM framework, F&O ban periods, operator activity warnings, budget/RBI policy patterns. Read before recommending any stock outside Nifty 50, and always for mid/small caps.
references/india-mechanics.md - — Unusual volume detection, bulk/block deal tracking, sector-stock divergence, historical pattern context, India VIX framework, gap risk/slippage management, portfolio heat rule. Read for every trade plan to add genuine analytical edge.
references/alpha-and-edge.md
references/- —— 完整的8参数评分标准,包含权重、阈值和展示格式。计算信心评分时阅读。
references/conviction-scoring.md - —— 相对强度排名、多时间框架验证、跳空分析、财报预警、行业轮动。进行深度分析或完整交易计划时阅读。
references/advanced-filters.md - —— 观察名单维护、风险仪表盘、仓位规模计算器、新闻/事件标记、Zerodha适配指令格式。用户要求组合管理或指令执行帮助时阅读。
references/workflow-tools.md - —— Kite MCP集成细节:可用工具、各技能模式下的使用方法、执行安全规则、用户设置说明。检测到Kite MCP工具或用户询问连接Zerodha账户时阅读。
references/kite-mcp.md - —— 涨跌幅限制、T+1结算、ASM/GSM框架、F&O限制期、庄家操控预警、预算/RBI政策交易模式。推荐Nifty 50以外的股票前,以及分析中/小盘股时必须阅读。
references/india-mechanics.md - —— 异常成交量检测、大额/大宗交易跟踪、行业-股票背离、历史形态背景、India VIX框架、跳空风险/滑点管理、组合总风险规则。每笔交易计划都需阅读以增加真正的分析优势。",
references/alpha-and-edge.md