Skip to main content Accueil Créateurs baggat236 ai-trading-skills canslim-screener
canslim-screener Screen US stocks using William O'Neil's CANSLIM growth stock methodology. Use when user requests CANSLIM stock screening, growth stock analysis, momentum stock identification, or wants to find stocks with strong earnings and price momentum following O'Neil's investment system.
Aller à l'installation Skills Marketplace Découvrez et explorez les compétences IA créées par la communauté.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Afficher les détails du prompt Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
npx skills add https://github.com/BaggaT236/AI-Trading-Skills --skill canslim-screenerLa commande reste sur une seule ligne. Faites défiler horizontalement pour la vérifier avant de la copier.
Vous préférez une copie locale ? Téléchargez les fichiers actuellement disponibles dans SkillsMP.
Télécharger Zip Téléchargement... Plus depuis ce dépôt Validate data quality in market analysis documents and blog articles before publication. Use when checking for price scale inconsistencies (ETF vs futures), instrument notation errors, date/day-of-week mismatches, allocation total errors, and unit mismatches. Supports English and Japanese content. Advisory mode -- flags issues as warnings for human review, not as blockers.
This skill retrieves upcoming earnings announcements for US stocks using the Financial Modeling Prep (FMP) API. Use this when the user requests earnings calendar data, wants to know which companies are reporting earnings in the upcoming week, or needs a weekly earnings review. The skill focuses on mid-cap and above companies (over $2B market cap) that have significant market impact, organizing the data by date and timing in a clean markdown table format. Supports multiple environments (CLI, Desktop, Web) with flexible API key management.
This skill should be used when analyzing market breadth charts, specifically the S&P 500 Breadth Index (200-Day MA based) and the US Stock Market Uptrend Stock Ratio charts. Use this skill when the user provides breadth chart images for analysis, requests market breadth assessment, positioning strategy recommendations, or wants to understand medium-term strategic and short-term tactical market outlook based on breadth indicators. Also works WITHOUT chart images by fetching CSV data directly from public sources. All analysis and output are conducted in English.
Métiers associés SOC
Basé sur la classification professionnelle SOC
Explorateur de fichiers
24 fichiers name canslim-screener description Screen US stocks using William O'Neil's CANSLIM growth stock methodology. Use when user requests CANSLIM stock screening, growth stock analysis, momentum stock identification, or wants to find stocks with strong earnings and price momentum following O'Neil's investment system.
CANSLIM Stock Screener - Phase 3 (Full CANSLIM)
Overview
This skill screens US stocks using William O'Neil's proven CANSLIM methodology, a systematic approach for identifying growth stocks with strong fundamentals and price momentum. CANSLIM analyzes 7 key components: C urrent Earnings, A nnual Growth, N ewness/New Highs, S upply/Demand, L eadership/RS Rank, I nstitutional Sponsorship, and M arket Direction.
Phase 3 implements all 7 of 7 components (C, A, N, S, L, I, M), representing 100% of the full methodology .
Two-Stage Approach:
Stage 1 (FMP API + Finviz) : Analyze stock universe with all 7 CANSLIM componentsStage 2 (Reporting) : Rank by composite score and generate actionable reports
Key Features:
Composite scoring (0-100 scale) with weighted components
Finviz fallback for institutional ownership data (automatic when FMP data incomplete)Progressive filtering to optimize API usage
JSON + Markdown output formats
Interpretation bands: Exceptional+ (90+), Exceptional (80-89), Strong (70-79), Above Average (60-69)
Bear market protection (M component gating)
Phase 3.1 Component Weights (Original O'Neil weights):
C (Current Earnings): 15%
A (Annual Growth): 20%
N (Newness): 15%
S (Supply/Demand): 15%
L (Leadership/RS Rank): 20% — multi-period weighted RS (3m/6m/12m vs configurable benchmark)
I (Institutional): 10%
M (Market Direction): 5% Weighted RS = 0.40 × rel_3m + 0.30 × rel_6m + 0.30 × rel_12m
Available periods are re-normalized when some are missing. Default benchmark is ^GSPC;
override with --rs-benchmark SPY/QQQ/IWM/....
Fallback hierarchy when multi-period data is incomplete:
No benchmark → weighted absolute stock performance + 20% penalty.
All multi-period windows missing but >=50 bars of price history → fall back to the
legacy 365-day full-window absolute return as the scoring input (20% penalty if no
benchmark).
<50 bars of price history → score=0 with error set.
Phase 4: FINVIZ Elite integration → 10x faster execution
When to Use This Skill
"Find CANSLIM stocks"
"Screen for growth stocks using O'Neil's method"
"Which stocks have strong earnings and momentum?"
"Identify stocks near 52-week highs with accelerating earnings"
"Run a CANSLIM screener on [sector/universe]"
User wants to identify multi-bagger candidates
User is looking for growth stocks with proven fundamentals
User wants systematic stock selection based on historical winners
User needs a ranked list of stocks meeting O'Neil's criteria
Value investing focus (use value-dividend-screener instead)
Income/dividend focus (use dividend-growth-pullback-screener instead)
Bear market conditions (M component will flag - consider raising cash)
Prerequisites
FMP API key (free tier: 250 calls/day, sufficient for 35 stocks; Starter tier $29.99/mo for 40+ stocks)
Python 3.9+
requests (FMP API calls)beautifulsoup4 (Finviz web scraping)lxml (HTML parsing) pip install requests beautifulsoup4 lxml
Output Output Directory: reports/ (default) or custom via --output-dir
canslim_screener_YYYY-MM-DD_HHMMSS.json - Structured data for programmatic usecanslim_screener_YYYY-MM-DD_HHMMSS.md - Human-readable report
Market Condition Summary (trend, M score, warnings)
Top N CANSLIM Candidates (ranked by composite score)
Component Breakdown for each stock (C, A, N, S, L, I, M scores with details)
Rating interpretation (Exceptional+/Exceptional/Strong/Above Average)
Quality warnings and data source notes
Summary statistics (rating distribution)
Exceptional+ (90-100): All components near-perfect, aggressive buyExceptional (80-89): Outstanding fundamentals + momentum, strong buyStrong (70-79): Solid across components, standard buyAbove Average (60-69): Meets thresholds with minor weaknesses, buy on pullback
Workflow
Step 1: Verify API Access and Requirements Check if user has FMP API key configured:
FMP API key (free tier: 250 calls/day, sufficient for 40 stocks)Python 3.9+ with required libraries:
requests (FMP API calls)beautifulsoup4 (Finviz web scraping)lxml (HTML parsing)
pip install requests beautifulsoup4 lxml
If API key is missing, guide user to:
Step 2: Determine Stock Universe Option A: Default Universe (Recommended)
Use top 40 S&P 500 stocks by market cap (predefined in script):
python3 skills/canslim-screener/scripts/screen_canslim.py
Option B: Custom Universe
User provides specific symbols or sector:
python3 skills/canslim-screener/scripts/screen_canslim.py \
--universe AAPL MSFT GOOGL AMZN NVDA META TSLA
Option C: Sector-Specific
User can provide sector-focused list (Technology, Healthcare, etc.)
API Budget Considerations (Phase 3):
40 stocks × 7 FMP calls/stock = 280 API calls
FMP: 7 calls/stock (profile, quote, income×2, historical_90d, historical_365d, institutional)
Finviz: ~1.8 calls/stock (institutional ownership fallback, 2s rate limit, not counted in FMP budget)
Market data (^GSPC quote, ^VIX quote, ^GSPC 52-week history): 3 FMP calls
Total: ~283 FMP calls per screening run (exceeds 250 free tier)
Recommendation : Use --max-candidates 35 for free tier (35 × 7 + 3 = 248 calls), or upgrade to FMP Starter tier ($29.99/mo, 750 calls/day) for full 40-stock screening
Step 3: Execute CANSLIM Screening Script Run the main screening script with appropriate parameters:
cd skills/canslim-screener/scripts
python3 screen_canslim.py --api-key $FMP_API_KEY
python3 screen_canslim.py \
--api-key $FMP_API_KEY \
--max-candidates 40 \
--top 20 \
--output-dir ../../../
python3 screen_canslim.py --rs-benchmark SPY
python3 screen_canslim.py --disable-rs
Script Workflow (Phase 3 - Full CANSLIM):
Market Direction (M) : Analyze S&P 500 trend vs 50-day EMA (using real historical data for accurate EMA)
If bear market detected (M=0), warn user to raise cash
S&P 500 Historical Data : Fetch 52-week data for M component EMA and L component RS calculationStock Analysis : For each stock, calculate:
C Component : Quarterly EPS/revenue growth (YoY)A Component : 3-year EPS CAGR and stabilityN Component : Distance from 52-week high, breakout detectionS Component : Volume-based accumulation/distribution (up-day vs down-day volume)L Component : 52-week Relative Strength vs S&P 500I Component : Institutional holder count + ownership % (with Finviz fallback)
Composite Scoring : Weighted average with all 7 component breakdownRanking : Sort by composite score (highest first)Reporting : Generate JSON + Markdown outputs Expected Execution Time (Phase 3):
40 stocks: ~2 minutes (additional 52-week history fetch per stock for L component)
Finviz fallback adds ~2 seconds per stock (rate limiting)
L component requires 365-day historical data for each stock
Finviz Fallback Behavior:
Triggers automatically when FMP sharesOutstanding unavailable
Scrapes institutional ownership % from Finviz.com (free, no API key)
Increases I component accuracy from 35/100 (partial data) to 60-100/100 (full data)
User sees: ✅ Using Finviz institutional ownership for NVDA: 68.3%
Step 4: Read and Parse Screening Results The script generates two output files:
canslim_screener_YYYY-MM-DD_HHMMSS.json - Structured datacanslim_screener_YYYY-MM-DD_HHMMSS.md - Human-readable report Read the Markdown report to identify top candidates:
ls -lt canslim_screener_*.md | head -1
cat canslim_screener_YYYY-MM-DD_HHMMSS.md
Report Structure (Phase 3 - Full CANSLIM):
Market Condition Summary (trend, M score, warnings)
Top N CANSLIM Candidates (ranked, N = --top parameter)
For each stock:
Composite Score and Rating (Exceptional+/Exceptional/Strong/etc.)
Component Breakdown (C, A, N, S, L, I, M scores with details)
Interpretation (rating description, guidance, weakest component)
Warnings (quality issues, market conditions, data source notes)
Summary Statistics (rating distribution)
Methodology note (Phase 3: 7 components, 100% coverage)
Component Details in Report:
S Component : "Up/Down Volume Ratio: 1.06 ✓ Accumulation"L Component (Phase 3.1) : "3m/6m/12m: +12.4%/+18.7%/+44.1% (rel +5.2%/+8.3%/+22.0%) | RS: 88 (Strong)"I Component : "6199 holders, 68.3% ownership ⭐ Superinvestor" A new Summary Table appears above the candidate list in Phase 3.1 reports, showing
rank, symbol, composite score, rating, RS rating, and RS percentile for quick scanning.
Step 5: Analyze Top Candidates and Provide Recommendations Review the top-ranked stocks and cross-reference with knowledge bases:
Reference Documents to Consult:
references/interpretation_guide.md - Understand rating bands and portfolio sizingreferences/canslim_methodology.md - Deep dive into component meanings (now includes S and I)references/scoring_system.md - Understand scoring formulas (Phase 3 weights) For Exceptional+ stocks (90-100 points) :
All components near-perfect (C≥85, A≥85, N≥85, S≥80, L≥85, I≥80, M≥80)
Guidance: Immediate buy, aggressive position sizing (15-20% of portfolio)
Example: "NVDA scores 97.2 - explosive quarterly earnings (100), strong 3-year growth (95), at new highs (98), volume accumulation (85), RS leader (92), strong institutional support (90), uptrend market (100)"
For Exceptional stocks (80-89 points) :
Outstanding fundamentals + strong momentum
Guidance: Strong buy, standard sizing (10-15% of portfolio)
For Strong stocks (70-79 points) :
Solid across all components, minor weaknesses
Guidance: Buy, standard sizing (8-12% of portfolio)
Phase 3 Example: "Stock scores 77.5 - strong earnings (85), solid growth (80), near high (70), accumulation (60), RS leader (75), good institutions (60), uptrend (90)"
For Above Average stocks (60-69 points) :
Meets thresholds, one component weak
Guidance: Buy on pullback, conservative sizing (5-8% of portfolio)
If M component = 0 (bear market detected), do NOT buy regardless of other scores
Guidance: Raise 80-100% cash, wait for market recovery
CANSLIM does not work in bear markets (3 out of 4 stocks follow market trend)
Step 6: Generate User-Facing Report Create a concise, actionable summary for the user:
# CANSLIM Stock Screening Results (Phase 3 - Full CANSLIM)
**Date:** YYYY-MM-DD
**Market Condition:** [Trend] - M Score: [X]/100
**Stocks Analyzed:** [N]
**Components:** C, A, N, S, L, I, M (7 of 7, 100% coverage)
## Market Summary
[2-3 sentences on current market environment based on M component]
[If bear market: WARNING - Consider raising cash allocation]
## Top 5 CANSLIM Candidates
### 1. [SYMBOL] - [Company Name] ⭐⭐⭐
**Score:** [X.X]/100 ([Rating])
**Price:** $[XXX.XX] | **Sector:** [Sector]
**Component Breakdown:**
- C (Earnings): [X]/100 - [EPS growth]% QoQ, [Revenue growth]% revenue
- A (Growth): [X]/100 - [CAGR]% 3yr EPS CAGR
- N (Newness): [X]/100 - [Distance]% from 52wk high
- S (Supply/Demand): [X]/100 - Up/Down Volume Ratio: [X.XX]
- L (Leadership): [X]/100 - 52wk: [+X.X]% ([+X.X]% vs S&P) RS: [XX]
- I (Institutional): [X]/100 - [N] holders, [X.X]% ownership [⭐ Superinvestor if present]
- M (Market): [X]/100 - [Trend]
**Interpretation:** [Rating description and guidance]
**Weakest Component:** [X] ([score])
**Data Source Note:** [If Finviz used: "Institutional data from Finviz"]
[Repeat for top 5 stocks]
## Investment Recommendations
**Immediate Buy List (90+ score):**
- [List stocks with exceptional+ ratings]
- Position sizing: 15-20% each
**Strong Buy List (80-89 score):**
- [List stocks with exceptional ratings]
- Position sizing: 10-15% each
**Watchlist (70-79 score):**
- [List stocks with strong ratings]
- Buy on pullback
## Risk Factors
- [Identify any quality warnings from components]
- [Market condition warnings]
- [Sector concentration risks if applicable]
- [Data source reliability notes if Finviz heavily used]
## Next Steps
1. Conduct detailed fundamental analysis on top 3 candidates
2. Check earnings calendars for upcoming reports
3. Review technical charts for entry timing
4. [If bear market: Wait for market recovery before deploying capital]
---
**Note:** This is Phase 3 (Full CANSLIM: C, A, N, S, L, I, M - 100% coverage).
Resources
Scripts Directory (scripts/)
screen_canslim.py - Main orchestrator script
Entry point for screening workflow
Handles argument parsing, API coordination, ranking, reporting
Usage: python3 screen_canslim.py --api-key KEY [options]
fmp_client.py - FMP API client wrapper
Rate limiting (0.3s between calls)
429 error handling with 60s retry
Session-based caching
Methods: get_income_statement(), get_quote(), get_historical_prices(), get_institutional_holders()
finviz_stock_client.py - Finviz web scraping client ← NEW
BeautifulSoup-based HTML parsing
Fetches institutional ownership % from Finviz.com
Rate limiting (2.0s between calls)
No API key required (free web scraping)
Methods: get_institutional_ownership(), get_stock_data()
Calculators (scripts/calculators/):
earnings_calculator.py - C component (Current Earnings)
Quarterly EPS/revenue growth (YoY)
Scoring: 50%+ = 100pts, 30-49% = 80pts, 18-29% = 60pts
growth_calculator.py - A component (Annual Growth)
3-year EPS CAGR calculation
Stability check (no negative growth years)
Scoring: 40%+ = 90pts, 30-39% = 70pts, 25-29% = 50pts
new_highs_calculator.py - N component (Newness)
Distance from 52-week high
Volume-confirmed breakout detection
Scoring: 5% of high + breakout = 100pts, 10% + breakout = 80pts
supply_demand_calculator.py - S component (Supply/Demand) ← NEW
Volume-based accumulation/distribution analysis
Up-day volume vs down-day volume ratio (60-day lookback)
Scoring: ratio ≥2.0 = 100pts, 1.5-2.0 = 80pts, 1.0-1.5 = 60pts
leadership_calculator.py - L component (Leadership/Relative Strength)
52-week stock performance vs S&P 500 benchmark
RS Rank estimation (1-99 scale, O'Neil style)
Scoring: RS 90+ outperforming market = 100pts, RS 80-89 = 80pts
institutional_calculator.py - I component (Institutional)
Institutional holder count (from FMP)
Ownership % (from FMP or Finviz fallback)
Superinvestor detection (Berkshire Hathaway, Baupost, etc.)
Scoring: 50-100 holders + 30-60% ownership = 100pts
market_calculator.py - M component (Market Direction)
S&P 500 vs 50-day EMA
VIX-adjusted scoring
Scoring: Strong uptrend = 100pts, Uptrend = 80pts, Bear market = 0pts
References Directory (references/)
references/canslim_methodology.md (27KB) - Complete CANSLIM explanation
All 7 components with O'Neil's original thresholds
S component (Volume accumulation/distribution) detailed explanation
L component (Leadership/Relative Strength) detailed explanation
I component (Institutional sponsorship) detailed explanation
Historical examples (AAPL 2009, NFLX 2013, TSLA 2019, NVDA 2023)
references/scoring_system.md (21KB) - Technical scoring specification (Phase 3)
Phase 3 component weights and formulas (all 7 components)
Interpretation bands (90-100, 80-89, etc.)
Minimum thresholds for all 7 components
Composite score calculation examples
references/fmp_api_endpoints.md (18KB) - API integration guide (Phase 3)
Required endpoints for all 7 components
L component: 52-week historical prices endpoint
Institutional holder endpoint documentation
Finviz fallback strategy explanation
Rate limiting strategy
Cost analysis (Phase 3: ~283 FMP calls for 40 stocks, exceeds 250 free tier)
references/interpretation_guide.md (18KB) - User guidance
Portfolio construction rules
Position sizing by rating
Entry/exit strategies
Bear market protection rules
Read references/canslim_methodology.md first to understand O'Neil's system (now includes S and I)
Consult references/interpretation_guide.md when analyzing results
Reference references/scoring_system.md if scores seem unexpected
Check references/fmp_api_endpoints.md for API troubleshooting or Finviz fallback issues
Troubleshooting
Issue 1: FMP API Rate Limit Exceeded ERROR: 429 Too Many Requests - Rate limit exceeded
Retrying in 60 seconds...
Running multiple screenings within short time window
Exceeding 250 calls/day (free tier limit)
Other applications using same API key
Wait and Retry : Script auto-retries after 60sReduce Universe : Use --max-candidates 30 to lower API usageCheck Daily Usage : Free tier resets at midnight UTCUpgrade Plan : FMP Starter ($29.99/month) provides 750 calls/day
Issue 2: Missing Required Libraries ERROR: required libraries not found. Install with: pip install beautifulsoup4 requests lxml
pip install requests beautifulsoup4 lxml
pip install beautifulsoup4
pip install requests
pip install lxml
Issue 3: Finviz Fallback Slow Execution Execution time: 2 minutes 30 seconds for 40 stocks (slower than expected)
Finviz rate limiting (2.0s per request)
All stocks triggering fallback due to FMP data gaps
Accept Delay : 1-2 minutes for 40 stocks is normal with Finviz fallbackMonitor Fallback Usage : Check logs for "Using Finviz institutional ownership" messagesReduce Rate Limit (advanced): Edit finviz_stock_client.py, change rate_limit_seconds=2.0 to 1.5 (risk: IP ban) Note: Finviz fallback adds ~2 seconds per stock but significantly improves I component accuracy (35 → 60-100 points).
Issue 4: Finviz Web Scraping Failure WARNING: Finviz request failed with status 403 for NVDA
⚠️ Using Finviz institutional ownership data - FMP shares outstanding unavailable. Finviz fallback also unavailable. Score reduced by 50%.
Finviz blocking scraping requests (User-Agent detection)
Rate limit exceeded (too many requests)
Network issues or Finviz downtime
Wait and Retry : Rate limit resets after a few minutesCheck Internet Connection : Verify network access to finviz.comFallback Accepted : Script continues with FMP holder count only (I score capped at 70/100)Manual Verification : Check Finviz website manually for blocked IP
Script never fails due to Finviz issues
Falls back to FMP holder count only
User sees quality warning in report
Issue 5: No Stocks Meet Minimum Thresholds ✓ Successfully analyzed 40 stocks
Top 5 Stocks:
1. AAPL - 58.3 (Average)
2. MSFT - 55.1 (Average)
...
Bear market conditions (M component low)
Selected universe lacks growth stocks
Market rotation away from growth
Check M Component : If M=0 (bear market), raise cash per CANSLIM rulesExpand Universe : Try different sectors or market cap rangesLower Expectations : Average scores (55-65) may still be actionable in weak marketsWait for Better Setup : CANSLIM works best in bull markets
Issue 6: Data Quality Warnings ⚠️ Revenue declining despite EPS growth (possible buyback distortion)
⚠️ Using Finviz institutional ownership data (68.3%) - FMP shares outstanding unavailable.
These are not errors - they are quality flags from calculators
Revenue warning: EPS growth may be from share buybacks, not organic growth
Finviz warning: Data source switched from FMP to Finviz (still accurate)
Review component details in full report
Cross-check with fundamental analysis
Adjust position sizing based on risk level
Finviz data is reliable - no action needed for data source warnings
Important Notes
Phase 3 Implementation Status This is Phase 3 implementing all 7 of 7 CANSLIM components:
✅ C (Current Earnings) - Implemented
✅ A (Annual Growth) - Implemented
✅ N (Newness) - Implemented
✅ S (Supply/Demand) - Implemented
✅ L (Leadership/RS Rank) - Implemented
✅ I (Institutional) - Implemented
✅ M (Market Direction) - Implemented
Composite scores represent 100% of full CANSLIM methodology
Uses original O'Neil component weights (C 15%, A 20%, N 15%, S 15%, L 20%, I 10%, M 5%)
L component (20% weight) is the largest individual factor alongside A, emphasizing relative strength leadership
M component uses real 50-day EMA from historical data (not fallback estimate)
Finviz Integration Benefits Automatic Fallback System:
When FMP API doesn't provide sharesOutstanding, Finviz automatically activates
Scrapes institutional ownership % from Finviz.com (free, no API key)
Improves I component accuracy from 35/100 (partial) to 60-100/100 (full)
FMP API (primary): Institutional holder count + shares outstanding calculationFinviz (fallback): Direct institutional ownership % from web pagePartial Data (last resort): Holder count only, 50% penalty applied
39/39 stocks successfully retrieved ownership % via Finviz (100% success rate)
Average execution time: 2.54 seconds per stock
No errors or IP blocks during testing
Future Enhancements
FINVIZ Elite integration for pre-screening
Execution time: 2 minutes → 10-15 seconds
FMP API usage reduction: 90%
Larger universe possible (100+ stocks)
Data Source Attribution
FMP API : Income statements, quotes, historical prices, key metrics, institutional holdersFinviz : Institutional ownership % (fallback), market dataMethodology : William O'Neil's "How to Make Money in Stocks" (4th edition)Scoring System : Adapted from IBD MarketSmith proprietary system
Disclaimer This screener is for educational and informational purposes only.
Not investment advice
Past performance does not guarantee future results
CANSLIM methodology works best in bull markets (M component confirms)
Conduct your own research and consult a financial advisor before making investment decisions
O'Neil's historical winners include AAPL (2009: +1,200%), NFLX (2013: +800%), but many stocks fail to perform
Version: Phase 3.1 (multi-period RS)
Last Updated: 2026-05-03
API Requirements: FMP API (free tier: up to 35 stocks; Starter tier recommended for 40 stocks) + BeautifulSoup/requests/lxml for Finviz
Execution Time: ~2 minutes for 40 stocks
Output Formats: JSON + Markdown (now includes Summary Table and schema_version: "3.1")
Components Implemented: C, A, N, S, L, I, M (7 of 7, 100% coverage)
Phase 3.1 additions: multi-period RS (3m/6m/12m), --rs-benchmark, --disable-rs,
new RS fields (rs_rating, rs_rank_percentile, rs_3m_return, rs_6m_return,
rs_12m_return, rs_benchmark, rs_benchmark_relative_return, rs_component_score,
benchmark_52w_performance).
