Notre avis
Crée un tableau de bord web interactif pour l'analyse technique d'indicateurs boursiers en utilisant Plotly Dash ou Streamlit.
Points forts
- Support de plusieurs types de tableaux de bord (mono-symbole, multi-symbole, multi-timeframe, scanner)
- Thème sombre intégré et sélecteurs de symboles, d'échange et d'intervalle
- Auto‑rafraîchissement en temps réel avec dcc.Interval ou st.rerun()
- Utilisation de templates prédéfinis pour accélérer le développement
Limites
- Nécessite une configuration préalable des règles de dépendances
- Les tableaux de bord ne sont pas déployés automatiquement
- Limitée aux indicateurs techniques de base sans personnalisation avancée
Utilisez cette compétence lorsque vous avez besoin d'un tableau de bord d'analyse technique visuel et interactif pour un ou plusieurs symboles boursiers avec des indicateurs courants.
Évitez cette compétence si vous avez besoin d'une analyse quantitative complexe ou d'un système de trading automatisé sans interface utilisateur.
Analyse de sécurité
SûrThe skill creates a local web dashboard for technical analysis using Plotly Dash or Streamlit. It involves file creation, running a local Python server, and loading environment variables from .env, but there are no instructions for data exfiltration, system destruction, or any dangerous operations. The Bash usage is limited to starting the application locally, which is standard and safe.
Aucun point d'attention détecté
Exemples
Build a single symbol dashboard for SBIN using Dash with EMA and RSI indicators.Create a multi-timeframe Streamlit dashboard for RELIANCE with 5m, 15m, 1h, and D timeframes.Create a scanner-dashboard type in Dash for a watchlist of 10+ NSE symbols.name: indicator-dashboard description: Build a web dashboard for technical indicator analysis using Plotly Dash or Streamlit. Supports single-symbol, multi-symbol, and multi-timeframe layouts with real-time refresh. argument-hint: "[type] [symbol]" allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
Create a web dashboard for interactive technical analysis using Plotly Dash or Streamlit.
Arguments
Parse $ARGUMENTS as: type symbol
$0= dashboard type. Default: single- Dash types:
single,multi-symbol,multi-timeframe,scanner-dashboard - Streamlit types:
streamlit-single,streamlit-multi,streamlit-scanner
- Dash types:
$1= symbol (e.g., SBIN, RELIANCE). Default: SBIN
If no arguments, ask the user what kind of dashboard they want and whether they prefer Dash or Streamlit.
Instructions
- Read the indicator-expert rules, especially:
rules/dashboard-patterns.md— Dash app patternsrules/streamlit-patterns.md— Streamlit app patternsrules/plotting.md— Chart patternsrules/data-fetching.md— Data loading
- Create
dashboards/{dashboard_name}/directory (on-demand) - Create
app.pyindashboards/{dashboard_name}/ - Use the matching template from
rules/assets/
Dashboard Requirements
All dashboards must include:
- Dark theme: Dash uses
dbc.themes.DARKLY; Streamlit uses[theme] base = "dark"or CSS injection - Symbol input: Text input or dropdown for symbol selection
- Exchange selector: NSE, BSE, NFO, NSE_INDEX
- Interval selector: 1m, 5m, 15m, 1h, D
- Indicator selectors: Checkboxes/multiselect for overlay and subplot indicators
- Interactive chart: Plotly chart with
template="plotly_dark",xaxis_type="category" - Stats display: Key metrics (LTP, Change, Volume, indicator values)
- Auto-refresh: Dash uses
dcc.Interval; Streamlit usesst.rerun()withtime.sleep() - Load
.envfrom project root viafind_dotenv()
Dash Dashboard Types
single — Single Symbol Dashboard (Dash)
- One symbol with configurable indicators
- Overlays: EMA, SMA, Bollinger, Supertrend, Ichimoku (checkboxes)
- Subplots: RSI, MACD, Stochastic, Volume, ADX, OBV (checkboxes)
- Stats panel: LTP, day change, volume, selected indicator values
- Template:
rules/assets/dashboard_basic/app.py
multi-symbol — Multi-Symbol Watchlist (Dash)
- 4-6 symbols in a grid layout
- Each cell shows candlestick + one overlay indicator
- Bottom row: RSI comparison across all symbols
- Symbol list editable via input
multi-timeframe — MTF Analysis (Dash)
- 4-panel grid: 5m, 15m, 1h, D for same symbol
- Same indicators computed on each timeframe
- Confluence summary: "3/4 timeframes bullish"
- Template:
rules/assets/dashboard_multi/app.py
scanner-dashboard — Live Scanner (Dash)
- Watchlist of 10+ symbols
- Table showing: Symbol, LTP, RSI, EMA trend, Signal
- Color-coded rows (green=bullish, red=bearish)
- Click symbol to show detailed chart
- Auto-refresh every 30 seconds
Streamlit Dashboard Types
streamlit-single — Single Symbol Dashboard (Streamlit)
- Sidebar: symbol, exchange, interval, overlay/subplot multiselect
st.plotly_chart()for interactive chartsst.metric()for LTP, Change, RSI, EMA stats- Auto-refresh via checkbox +
st.rerun() - Template:
rules/assets/streamlit_basic/app.py
streamlit-multi — MTF Analysis (Streamlit)
- 2x2 grid via
st.columns(2)for 4 timeframes - Candlestick + EMA overlay per timeframe
- Confluence summary with
st.success()/st.error()/st.warning() st.metric()cards for each timeframe trend- Template:
rules/assets/streamlit_multi/app.py
streamlit-scanner — Scanner Dashboard (Streamlit)
- Sidebar: scan type selector, run button
st.progress()during scanst.dataframe()for results tablest.download_button()for CSV export
Running the Dashboard
After creating the app, provide instructions:
Dash:
cd dashboards/{dashboard_name}
python app.py
# Open http://127.0.0.1:8050 in browser
Streamlit:
cd dashboards/{dashboard_name}
streamlit run app.py
# Open http://localhost:8501 in browser
Example Usage
/indicator-dashboard single SBIN
/indicator-dashboard multi-timeframe RELIANCE
/indicator-dashboard scanner-dashboard
/indicator-dashboard streamlit-single SBIN
/indicator-dashboard streamlit-multi RELIANCE
/indicator-dashboard streamlit-scanner
Ingénierie de Prompts
Data & IA
Bonnes pratiques et templates de prompt engineering pour maximiser les résultats IA.
Visualisation de Données
Data & IA
Génère des visualisations de données et graphiques adaptés à vos données.
Architecture RAG
Data & IA
Guide de configuration d'architectures RAG (Retrieval-Augmented Generation).