agentleFS
ShubhamkumartrichyShubhamkumartrichy/claude-trading-skills-india/skills/breakout-planner/SKILL.md

breakout-planner

Skill64 starsChanged 17 days agoMITView on GitHub

Stage-2 breakout setups on NSE/BSE stocks: a trend-template check (rising 200 DMA, 50>150>200 stack, distance from 52-week high/low, relative strength vs NIFTY), base detection (volatility contraction with successively tighter pullbacks, or a flat base) with pivot, depth, duration, volume dry-up and a grade, then an order-ready plan: trigger, chase limit, capped stop, size, Indian charges, R targets, GTT/SL templates, circuit-band and NSE tick-size checks. breakout.py scans a folder of CSVs or yfinance symbols and ranks them, offline-capable.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run ../market-regime/scripts/regime.py status first (see market-regime).

    line 106
  2. Very specific:

    Run scan on the watchlist, then check the top names and confirm the base on the chart (technical-analysis, multi-timeframe-analysis).

    line 108
  3. Very specific:

    Run plan for actionable names.

    line 110
  4. Very specific:

    After entry, hand trade management to trailing-stops: breakeven at 1R, trail from 2R.

    line 112

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Breakout Planner (NSE/BSE)

Buy strength only where three things line up: a stock already in a Stage-2 uptrend, a tight base whose pullbacks shrink as supply dries up, and a pivot you can define to the tick. The script does the detection and arithmetic. You add regime, news and judgement.

Commands

# One stock: trend template + liquidity + base + pivot (add a NIFTY CSV for relative strength)
python3 scripts/breakout.py check --csv polycab.csv --benchmark-csv nifty.csv
python3 scripts/breakout.py check --symbol POLYCAB --benchmark NIFTY          # yfinance
# Plan from the chart (pivot, last low and ATR are read from the data)
python3 scripts/breakout.py plan --csv polycab.csv --capital 1000000 --risk-pct 1 --band 0
# Plan from numbers you already have (no data needed)
python3 scripts/breakout.py plan --pivot 1520 --contraction-low 1462 --prev-close 1498 \
    --capital 500000 --band 10 --segment mtf --leverage 2 --broker zerodha
# Rank many: one CSV per symbol (file name = symbol), or a yfinance list
python3 scripts/breakout.py scan --csv-dir data/ --benchmark-csv data/nifty.csv --top 15
python3 scripts/breakout.py scan --symbols DIXON,KAYNES,TRENT,POLYCAB --benchmark NIFTY

Inputs follow technical-analysis: --csv, --candles (broker-MCP JSON) or --symbol. Use 260+ daily bars. Every command takes --json. Bad input exits 2 with a one-line error.

Reading check

Trend template (all must pass for "Stage 2"): close above SMA50/150/200; SMA50 > SMA150 > SMA200; SMA200 higher than 22 sessions ago; within 25% of the 52-week high; at least 30% above the 52-week low; with a benchmark, beating it over both 3 and 6 months. Use NIFTY for large caps and NIFTY Midcap/Smallcap or the sector index for the rest. Five of six is "watch", not "buy".

Base. The base starts at the highest high of the last 150 bars. Swings are traced with a threshold of max(3%, 1.25 × ATR%). A bounce that fails into a lower low stays part of the same pullback; a small dip on the way to a higher high is ignored.

  • VCP: 2+ pullbacks, each shallower than the previous (≤0.85x), last one ≤15% (ideally ≤6%), first ≤50%. Pivot = high of the last pullback.
  • Flat base: whole base ≤15% deep and 5+ weeks long when the pullbacks don't tighten. Pivot = base high.
  • Loose/unclassified: neither. Grade D. Wait for a proper base.
  • Volume dry-up = volume in the last pullback ÷ the 50 bars before it. Below 0.6 is strong; above 1.0 means supply.
  • Grade A ≥80, B ≥60, C ≥40, D below. State: pre-breakout, breakout zone (≤ chase above pivot), extended, deep below pivot, or failed (closed under the last low).
  • Cup-with-handle is not detected separately. Most read as a VCP or loose base, so confirm on the chart.

Reading plan

ItemRule (defaults)
Triggerpivot + 0.2%, rounded up to the tick
Chase limitpivot + 3% (--max-chase-pct; warns above 5%). Size is computed at this worst fill
Stoplast contraction low − 1% (--stop-method atr = trigger − 2×ATR), capped at 8% below the limit
Sizeposition-sizing logic (risk.py): risk % of capital incl. costs, 20% single-stock cap; MTF divides the risk budget by the leverage
Chargestrading-costs (charges.py) round trip at the stop and at each target, delivery or MTF with interest
Targets2R and 3R from the worst fill (--targets 2,3,5)

A capped stop sits inside the base, so the base is too loose for your chase. Prefer to skip it or buy closer to the pivot.

India specifics

  • Order types. Exchange stop orders (SL / SL-M) are DAY-only and lapse at 15:30, and the pre-open session takes no stop orders. For a multi-day wait use a broker GTT (trigger = entry trigger, limit = chase limit). After the fill, protect it with a GTT OCO (stop + target). Shares settle T+1, so on the buy day some brokers need a DAY SL-sell until the shares show in holdings.
  • Tick size follows NSE slabs effective 15-Apr-2025: below ₹250 ₹0.01; ₹250–1,000 ₹0.05; ₹1,000–5,000 ₹0.10; ₹5,000–10,000 ₹0.50; ₹10,000–20,000 ₹1; above ₹20,000 ₹5. NSE reassigns slabs monthly from a reference close, and BSE can differ, so pass --tick if the broker rejects a price.
  • Circuit bands. --band 2|5|10|20 with --prev-close (taken from the data's last close) warns when the trigger or chase limit is beyond today's upper circuit. 2%/5% bands mark surveillance, SME or illiquid names, where breakouts lock at the upper circuit and exits lock at the lower one. F&O stocks have no fixed circuit (--band 0): the exchange widens their dynamic band in steps.
  • Surveillance. ASM/GSM stocks and trade-for-trade (BE/T2T) stocks have no intraday or BTST, need 100% upfront payment and usually can't be funded with MTF. Check NSE's lists or pass --surveillance asm|gsm|t2t.
  • Liquidity. The default floor is ₹5 crore average daily traded value (--min-turnover-cr). Scans push illiquid names to the bottom. Skip SME and operator-style stocks: a vertical base on rising volume with no news or delivery is a red flag.
  • Delivery % on the breakout day should match or beat its recent average (NSE security-wise delivery data). A high-volume breakout on low delivery is mostly intraday churn.
  • Results season. A pivot within ~5 sessions of quarterly results is a coin toss with a 5–15% gap. Wait for the number or size down. The plan reports the worst 60-bar gap-down and sizes the gap-through loss for it.

Workflow

  1. Run ../market-regime/scripts/regime.py status first (see market-regime). Take new breakouts only when the regime allows long exposure, and respect its position cap.
  2. Run scan on the watchlist, then check the top names and confirm the base on the chart (technical-analysis, multi-timeframe-analysis).
  3. Run plan for actionable names. Cross-check the stop with stop-loss-strategies and the size with position-sizing.
  4. After entry, hand trade management to trailing-stops: breakeven at 1R, trail from 2R. Exit fast if price closes back under the pivot on heavy volume within a few sessions.

Output

SYMBOL — date, close | trend 6/6 | base VCP A | pre-breakout (−2.1% vs pivot)
T1 20% → T2 10% → T3 5% | dry-up 0.40x | pivot ₹P | last low ₹L
Trigger ₹T | chase ≤ ₹C | stop ₹S (x% , capped?) | qty N (₹value, y% of capital) | risk ₹R
Targets 2R ₹.. / 3R ₹.. (net of charges) | band check | tick
Orders: GTT buy (trigger/limit) → GTT OCO (stop/target); DAY SL alternative
Checklist: regime, surveillance, results date, delivery %, liquidity

Educational analysis, not investment advice.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.