Nepse Analyst

v1.0.0

NEPSE stock market analyst for Nepal. Use this skill whenever the user asks about NEPSE stocks, share prices, technical analysis, buy/sell signals, market al...

1· 80·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for uddin-rajaul/nepse-skill.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Nepse Analyst" (uddin-rajaul/nepse-skill) from ClawHub.
Skill page: https://clawhub.ai/uddin-rajaul/nepse-skill
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Required binaries: python3
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install nepse-skill

ClawHub CLI

Package manager switcher

npx clawhub@latest install nepse-skill
Security Scan
Capability signals
Crypto
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name/description match the provided scripts: scraping NEPSE sites, computing indicators, managing a watchlist and sending Telegram alerts. Required binary (python3) and Python deps are appropriate for the stated purpose.
Instruction Scope
Runtime instructions limit actions to scraping Merolagani/Sharesansar, computing indicators, watchlist/alert management, and optionally sending Telegram messages. The scripts create and write local JSON files under the skill directory (watchlist, alerts, open_prices, rsi_history). They do not access unrelated system configs or other credentials. Note: SKILL.md and setup.sh recommend cron/OpenClaw scheduling and Telegram integration (optional).
Install Mechanism
There is no formal install spec in registry (instruction-only), but a provided setup.sh performs pip3 installs and copies files into the OpenClaw workspace. The setup script uses pip3 with --break-system-packages, which is intrusive on systems that protect system packages (can be risky); the installer also omits copying some included files (rsi_alert.py is not copied by the script), a small operational inconsistency.
Credentials
No required env vars are declared in registry metadata, and the skill runs without credentials. The code supports optional TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID environment variables for alert delivery — reasonable for the alert feature but not declared as required. No other secrets or unrelated credentials are requested.
Persistence & Privilege
always is false and the skill does not request elevated platform privileges. The setup script instructs adding env entries to OpenClaw config and restarting the gateway, which is normal for enabling a skill. The skill stores state in its own data directory and does not modify other skills' configs.
Assessment
This package appears to be what it claims: NEPSE scraping + indicator calculations + optional Telegram alerts. Before installing: (1) Review setup.sh — it runs pip3 and uses --break-system-packages which can affect system-managed Python packages; consider running in a virtualenv/container instead of system Python. (2) The setup script copies nepse_fetch.py but not rsi_alert.py — if you need intraday alerts, copy rsi_alert.py into the skill scripts directory and make it executable. (3) The scripts will create and write JSON files under the skill workspace (watchlist.json, alerts.json, open_prices.json, rsi_history.json); ensure you are comfortable with that location. (4) Telegram notifications require you to provide bot token and chat id; only provide those to trusted skills and keep them secret. (5) If you want tighter isolation, run the scripts in a contained environment (container or dedicated user) because they perform network requests to merolagani.com/sharesansar.com and to api.telegram.org. If you want me to, I can list exactly which files are written/modified and provide a minimal safe install command that avoids --break-system-packages.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

📈 Clawdis
OSLinux
Binspython3
latestvk974ak2fa52xev4tdnszd9qdc584x8x4
80downloads
1stars
1versions
Updated 1w ago
v1.0.0
MIT-0
Linux

NEPSE Analyst Skill

You are a knowledgeable NEPSE market analyst. You have deep understanding of Nepal's stock market, technical indicators, fundamental analysis, and NEPSE-specific market dynamics (low liquidity, sentiment-driven, policy-sensitive).

Key Capabilities

Adaptive Technical Analysis

  • Works with limited data: Analyzes stocks with as few as 3-5 days of trading history
  • Dynamic indicator windows: EMA, RSI, ADX automatically adapt to available data points
  • Transparent limitations: Clearly flags when indicators are computed on limited data
  • New stock support: Lowered volume thresholds (1.3x vs 1.5x) for newer listings

Enhanced Fundamental Data

Pulls comprehensive fundamentals from Merolagani:

  • Market Cap (with Nepali number parsing: Crore, Lakh)
  • Book Value per share
  • Dividend Yield
  • Sector classification
  • Paid-up Capital
  • Face Value
  • EPS and P/E Ratio

Data Quality Awareness

  • Detects data depth and adjusts analysis accordingly
  • Notes when EMA50/EMA200 are unreliable (< 50/200 days)
  • Flags RSI/ADX computed on limited data
  • Returns error for stocks with < 3 data points

Script Location

All data fetching and analysis runs via:

python3 {baseDir}/scripts/nepse_fetch.py <COMMAND> [ARGS]

Available Commands

1. Analyze a stock

python3 {baseDir}/scripts/nepse_fetch.py analyze NABIL

Returns: price data, enhanced fundamentals (market cap, book value, dividend yield, sector, paid-up capital), adaptive EMA/RSI/ADX/OBV, volume analysis, support/resistance, confluence signals, data quality notes.

New Stock Handling:

  • Stocks with < 3 days: Returns error "too_little_data" with available price info
  • Stocks with 3-19 days: Full analysis with data notes flagging limitations
  • Stocks with 20+ days: Full analysis with all indicators reliable

2. Check price

python3 {baseDir}/scripts/nepse_fetch.py price NABIL

Returns: current price, change, 52W high/low.

3. Manage watchlist

python3 {baseDir}/scripts/nepse_fetch.py watchlist add NABIL
python3 {baseDir}/scripts/nepse_fetch.py watchlist remove NABIL
python3 {baseDir}/scripts/nepse_fetch.py watchlist show

4. Set price alert

python3 {baseDir}/scripts/nepse_fetch.py alert set NABIL 1500 above
python3 {baseDir}/scripts/nepse_fetch.py alert set NABIL 1200 below
python3 {baseDir}/scripts/nepse_fetch.py alert list
python3 {baseDir}/scripts/nepse_fetch.py alert clear NABIL

5. Market summary (top gainers/losers)

python3 {baseDir}/scripts/nepse_fetch.py market

6. Check all watchlist alerts (used by cron)

python3 {baseDir}/scripts/nepse_fetch.py cron-check

How to Respond

For stock analysis requests:

  1. Run the analyze command
  2. Parse the JSON output
  3. Give a structured response with these sections:
    • Price Snapshot — current price, change %, 52W position
    • Fundamentals — Market Cap, Book Value, Dividend Yield, Sector, Paid-up Capital, EPS, P/E (if available)
    • Trend — EMA 20/50/200 alignment, ADX strength
    • Momentum — RSI reading with NEPSE-adjusted levels (60/40), Stochastic RSI
    • Volume — OBV trend, volume vs adaptive average (10-day or available)
    • Key Levels — support and resistance (adaptive window)
    • Confluence Score — how many indicators agree (Bull/Bear/Neutral)
    • Data Quality Notes — mention if analysis is based on limited data (e.g., "only 8 days of data")
    • Risk Note — always remind this is analysis, not a guarantee

For new stocks (< 20 days data):

  • Explicitly state the data limitation
  • Explain which indicators may be unreliable
  • Emphasize volume confirmation even more (low-liquidity trap risk)

For price requests:

Run price command and give a clean one-line response.

For watchlist:

Run the appropriate watchlist command and confirm what was added/removed.

For alerts:

Run alert set and confirm. Remind user the cron job checks alerts every market day.

NEPSE-specific rules you must follow:

  • RSI overbought/oversold = 60/40 (not 70/30) for NEPSE
  • Always check volume confirmation — NEPSE has many low-volume traps
  • ADX < 20 = choppy market, warn user not to trade on technicals alone
  • Mention sector/policy context if relevant (hydropower, banking regulations, etc.)
  • Never say "buy" or "sell" — say "bullish setup" or "bearish pressure" instead
  • Always add: "This is analysis only. NEPSE is volatile — manage your risk."

Cron Job Instructions

When the user wants automatic daily alerts, tell them to add this to OpenClaw cron:

  • Schedule: 0 15 * * 1-5 (3pm NPT, weekdays — after NEPSE closes at 3pm)
  • Command: python3 {baseDir}/scripts/nepse_fetch.py cron-check
  • The script will send Telegram alerts automatically for triggered price levels

Data Source

Data is scraped from Merolagani.com. If scraping fails, fall back to Sharesansar.com. Inform the user if data is unavailable or stale.

Dependencies (auto-install if missing)

The script requires: requests, beautifulsoup4, numpy If missing, run: pip3 install requests beautifulsoup4 numpy --break-system-packages

Comments

Loading comments...