agentleFS
ShubhamkumartrichyShubhamkumartrichy/claude-trading-skills-india/skills/technical-analysis/SKILL.md

technical-analysis

Skill64 starsChanged 16 days agoMITView on GitHub

Systematic chart reading for NSE/BSE stocks and indices: trend and moving-average alignment, swing structure, support/resistance clusters, unfilled gaps, volume and delivery %, an indicator dashboard (RSI, MACD, Bollinger, ADX, Stochastic), ATR stops, circuit and gap-risk checks, and probabilistic scenarios. Ships a zero-dependency engine (ta.py) that reads CSV, broker-MCP candles or fetches via yfinance, verified against the `ta` library.

  • Installs packages

11 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use fundamental-analysis.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • iceyccSkill

    fundamental-analysis

    股票基本面分析和选股工具集,支持获取财务指标(PE、PB、ROE、EPS、BPS、资产负债率等)、执行基本面选股策略(格雷厄姆 PB 选股、多因子选股)。用于分析公司基本面、筛选优质股票、执行价值投资策略。当需要获取股票财务指标、进行基本面选股、或执行价值投资策略时使用。

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the chart systematically and report every indicator, not only the ones that agree with a bias.

    line 17
  2. Distinctive:

    Use at least 250 daily bars so SMA200 and 52-week levels exist.

    line 34
  3. Distinctive:

    Always give 2–3 scenarios whose probabilities sum to ~100%.

    line 108

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “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

Technical Analysis (NSE/BSE)

Read the chart systematically and report every indicator, not only the ones that agree with a bias. Let the script do the arithmetic. Your job is interpretation and context.

1. Get candles

Pick the first option that works:

# A. Broker MCP: save the candle response to a JSON file, then
python3 scripts/ta.py summary --candles candles.json --symbol RELIANCE
# B. yfinance (pip install yfinance). NSE stocks auto-map to .NS; indices: NIFTY, BANKNIFTY, SENSEX, INDIAVIX, NIFTYIT …
python3 scripts/ta.py summary --symbol RELIANCE              # --exchange BSE for .BO
python3 scripts/ta.py fetch --symbol NIFTY --period 5y --out nifty.csv
# C. User/exported CSV (Date, Open, High, Low, Close, Volume, any case)
python3 scripts/ta.py summary --csv reliance.csv

Use at least 250 daily bars so SMA200 and 52-week levels exist. yfinance prices are split-adjusted but not dividend-adjusted, which matches broker charts. Check the last bar's date. The script warns when data is stale. Intraday bars from yfinance are limited (60 days for 15m, 730 days for 1h).

Other sub-commands: levels, divergence, fib, mtf, indicators --rows 15; add --json for raw output.

2. Trend

  • Structure: higher highs and higher lows is an uptrend; lower highs and lower lows is a downtrend; anything else is a range. The script reports it from swing points (--swing 5 = 5 bars each side).
  • MA alignment: price > SMA20 > SMA50 > SMA200 is a strong uptrend, the reverse a strong downtrend, and mixed means transition or chop.
  • Golden or death cross (SMA50 vs SMA200): the script reports recent crosses. It is a lagging confirmation, not an entry.
  • Relative strength: compare the stock to NIFTY and to its sector index (NIFTYIT, NIFTYBANK, NIFTYAUTO, …). Leaders in a weak market and laggards in a strong one are different trades.

3. Support and resistance

  • Swing-point clusters within ~1.5% (the script counts touches). Two or more touches is meaningful.
  • 52-week high/low and round numbers (₹1,000, ₹2,500; NIFTY 25,000; BANKNIFTY 55,000) act as magnets. Indian retail and option writers anchor heavily on round strikes.
  • Unfilled gaps: gap-up zones act as support and gap-down zones as resistance. Overnight gaps are frequent because of GIFT Nifty and US cues.
  • Dynamic levels: SMA20 (short), SMA50 (medium), SMA200 (major).
  • For indices and F&O stocks: the largest call-OI strike tends to cap rallies and the largest put-OI strike tends to hold dips into expiry (see fno-options-analysis).

4. Volume and delivery

PriceVolumeRead
UpRisingHealthy advance
UpFallingWeak rally, suspect
DownRisingDistribution, trend acceleration
DownFallingSelling exhaustion, bounce possible
Spike at support—Capitulation or absorption
Spike at resistance—Breakout attempt or distribution

More than 1.5x the 20-day average is significant. Delivery % is an India-specific edge. NSE publishes the delivery quantity as a percentage of traded quantity per stock per day (security-wise delivery data, or the broker's "delivery %" field). A rising price with a rising delivery % means real buyers taking delivery. A rising price with a low delivery % (below about 30% in large caps) is intraday churn.

5. Indicator dashboard

The script grades each indicator bullish, bearish or neutral and counts them:

IndicatorBullishBearishNeutral
RSI(14)>50 and rising; rising out of <30<50 and falling; falling from >7040–60 flat
MACD(12,26,9)histogram >0 and growinghistogram <0 and growingnear zero
Bollinger(20,2)%B 0.5–1 (upper half)%B 0–0.5squeeze or walking the band
ADX(14)>25 with +DI > −DI>25 with −DI > +DI<20 (no trend)
Stochastic(14,3,3)%K crosses up below 20–30%K crosses down above 70–80mid-range
MA alignmentstacked upstacked downmixed

Indicators are correlated (RSI, Stochastic and MACD all measure momentum). Five bullish signals is not five independent confirmations. Weight structure and volume above the count.

6. India-specific checks (script prints these)

  • Circuit hit: a close at the day's high or low exactly at a band (2/5/10/20%) means one-sided liquidity. The next day can gap again, and exits may be impossible in a lower circuit.
  • Gap statistics: average and maximum overnight gap over 60 bars. If the stop is closer than the typical gap, it is decoration, not protection.
  • Event days: quarterly results (the stock can move 5–15%), ex-dividend (mechanical price drop), bonus/split (check adjustment), index inclusion or exclusion, and F&O ban entry or exit.
  • Closing price is a 30-minute VWAP (15:00–15:30) on NSE, so "closing basis" stops use that number, not the last tick.

7. Scenarios

Always give 2–3 scenarios whose probabilities sum to ~100%. Tilt toward the scenario with better structure + volume + higher-timeframe agreement. Keep near-equal odds when evidence conflicts.

A — Bullish (xx%): trigger [daily close above ₹X on >1.5x volume] → targets ₹T1/₹T2; invalid below ₹S
B — Bearish (xx%): trigger [close below ₹S] → targets ₹…; invalid above ₹…
C — Range (xx%): ₹L–₹H for ~N sessions; breakout tell: [volume/delivery/ADX rising above 20]

Output

Lead with the verdict and the data timestamp. Then trend, levels (nearest 2–3 each side with touches), dashboard counts, volume and delivery, divergences, ATR and stop options, India checks, and scenarios. Hand off to position-sizing and stop-loss-strategies when the user wants a trade plan.

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.