★ ★ ★LIVE97 COINS ANALYZED BY CLAUDE MCP·VIEW SKILLS →★ ★ ★
MCP
claude.analyze("BTCUSDT")→ [buy] signal detectedmcp.draw_levels(resistance=98000)→ [drawn] on chartclaude.scan_watchlist(coins=100)→ [42] buy signals foundmcp.connect("tradingview")→ [connected] daily TFclaude.detect_sr("SOLUSDT")→ [S: $165] [R: $185]claude.analyze("ETHUSDT")→ [neutral] consolidatingmcp.screenshot_chart("BNBUSDT")→ [captured] analyzing...claude.score_quality("XRPUSDT")→ score: 78/100mcp.set_timeframe("1D")→ [ok] chart updatedclaude.analyze("BTCUSDT")→ [buy] signal detectedmcp.draw_levels(resistance=98000)→ [drawn] on chartclaude.scan_watchlist(coins=100)→ [42] buy signals foundmcp.connect("tradingview")→ [connected] daily TFclaude.detect_sr("SOLUSDT")→ [S: $165] [R: $185]claude.analyze("ETHUSDT")→ [neutral] consolidatingmcp.screenshot_chart("BNBUSDT")→ [captured] analyzing...claude.score_quality("XRPUSDT")→ score: 78/100mcp.set_timeframe("1D")→ [ok] chart updated
BACK TO CLAUDE SKILLS
◈

analyze_chart_screenshot

Claude AIClaude APIVision
INSTALLATION
$python -c "exec(open('analyze_chart_screenshot.py').read())"
#or paste directly into your Claude Code terminal
[ ABOUT ]

Passes a base64-encoded chart screenshot to Claude using the vision modality, requesting identification of key support and resistance levels from the chart image. The image content block is structured per the Anthropic API spec (`type: image`, `source.type: base64`) so no external image hosting is required.

Vision-based chart analysis captures information that numerical data cannot — drawn trend lines, volume bars, candlestick patterns, and chart annotations are all visible in screenshots but absent from raw OHLCV arrays. Use this function as a complement to numerical analysis: compute S/R levels programmatically, then pass the chart screenshot to Claude to verify and identify any levels the algorithm missed.

[ USAGE EXAMPLE ]
example.py
img = screenshot_chart(client)
levels = analyze_chart_image(img["data"])
# → "Key support at $94,200 and $91,500. Resistance at $98,000."
[ FULL CODE ]
analyze_chart_screenshot.py
def analyze_chart_image(img_b64: str): msg = client.messages.create(model="claude-sonnet-4-5", max_tokens=500, messages=[{"role":"user","content":[{"type":"image","source":{"type":"base64","media_type":"image/png","data":img_b64}},{"type":"text","text":"List key S&R levels"}]}]) return msg.content[0].text
[ METADATA ]
CategoryClaude AI
ComplexityBeginner
LanguagePython 3.10+
TagsClaude API, Vision
[ ASK CLAUDE ]

Ask AI about this skill before installing it.

[ RELATED SKILLS ]