MarketMind AI — MCP Hackathon Submission

submissions/mcp-hackathon/marketmind-ai/RIGHTS.md

@@ -0,0 +1,22 @@
+# Submission rights declaration
+
+Project: `MarketMind AI`
+Submission slug: `marketmind-ai`
+Submitter: `gurusankar55`
+Date: `2026-09-10`
+
+The submitter confirms that they own, or have sufficient authorization for, the source code, dependencies, service, data, branding, and other materials submitted in this pull request.
+
+Subject to the official program terms, the submitter authorizes X-Agent to retain, reproduce, audit, test, archive, and publish the submitted program artifact for judging, fraud prevention, dispute handling, ecosystem submission, and post-award accountability.
+
+Third-party components and their licenses:
+- FastAPI — MIT License
+- Pandas — BSD-3-Clause License
+- Requests — Apache License 2.0
+- Uvicorn — BSD-3-Clause License
+- Kraken public market-data API
+
+Exceptions or restrictions:
+- No known restrictions.
+- Market data is obtained from a public third-party API.
+- The service is provided for informational market intelligence and does not execute trades or manage wallets.

submissions/mcp-hackathon/marketmind-ai/SUBMISSION.md

@@ -0,0 +1,58 @@
+# MarketMind AI
+
+## Capability
+
+- One-line description: Agent-callable crypto market intelligence API that returns market indicators and a structured market assessment for supported trading pairs.
+- Who it helps: AI agents, developers, and users who need structured crypto market research data through a simple HTTP API.
+- Capability boundary: Market research and analysis only. It does not execute trades, place orders, manage wallets, or provide security/audit services.
+
+## Live API
+
+- API base URL: https://marketmind-ai-js97.onrender.com
+- Health-check URL: https://marketmind-ai-js97.onrender.com/health
+- Authentication: none
+- Rate limits / known limits: Public Render free service may spin down after inactivity. Market data availability depends on the public Kraken API and supported trading pairs.
+- API contract: GET /v1/market-intelligence/{symbol}
+
+Example:
+GET /v1/market-intelligence/BTCUSDT
+
+## Source and reproducibility
+
+- Source repository: https://github.com/gurusankar55/MarketMind-AI
+- Review commit: `2578c4097be83add08543464b05c628b07425b79`
+- Source submitted in this PR: `source/`
+- Run tests: `python -c "from candles import get_candles; print(get_candles('BTCUSDT').tail())"`
+- Run locally: `cd app && uvicorn main:app --reload`
+- Deploy: Render Web Service using `pip install -r requirements.txt` and `cd app && uvicorn main:app --host 0.0.0.0 --port $PORT`
+- Version binding:
+
+GET /health
+
+{"status":"ok","commit":"2578c4097be83add08543464b05c628b07425b79"}
+
+GET /.well-known/xagent-verification.json
+
+{"schemaVersion":1,"slug":"marketmind-ai","commit":"2578c4097be83add08543464b05c628b07425b79"}
+
+## Verification
+
+The reproducible call instructions and example responses are in `verification/README.md`.
+
+- Health-check result: HTTP 200 with status `ok` and the exact deployed review commit.
+- Capability call: GET /v1/market-intelligence/BTCUSDT returns current market price, EMA20, EMA50, EMA200, RSI14, MACD, MACD signal, and a market assessment score/regime.
+- Expected error behavior: Unsupported or invalid symbols may return an HTTP error. The API does not require authentication.
+
+## Security and data handling
+
+- Data collected: Public cryptocurrency market candle data.
+- Purpose and retention: Data is used only to calculate market indicators and assessment for the API response. No user account data is stored.
+- Third parties / outbound network calls: Public Kraken market-data API.
+- Secrets: No secrets are committed. No authentication credentials are required.
+- Known risks / restrictions: Market data can change rapidly and should be treated as informational market intelligence, not guaranteed financial advice or trade execution.
+
+## Support
+
+- Team / builder: gurusankar55
+- Contact: GitHub repository issues
+- License / rights: Project source is submitted by the builder for hackathon review and archival.

submissions/mcp-hackathon/marketmind-ai/source/.gitignore

@@ -0,0 +1 @@
+__pycache__/ 

submissions/mcp-hackathon/marketmind-ai/source/README.md

@@ -0,0 +1,41 @@
+# MarketMind AI
+
+Agent-callable crypto market intelligence API.
+
+## Features
+
+- Public crypto market candle data
+- EMA 20, 50, 200
+- RSI 14
+- MACD
+- Market assessment score
+- Simple JSON API for AI agents
+- X-Agent verification endpoints
+
+## API
+
+GET `/v1/market-intelligence/{symbol}`
+
+Examples:
+
+`/v1/market-intelligence/BTCUSDT`
+
+`/v1/market-intelligence/ETHUSDT`
+
+`/v1/market-intelligence/SOLUSDT`
+
+## Health
+
+GET `/health`
+
+## X-Agent Verification
+
+GET `/.well-known/xagent-verification.json`
+
+## Tech Stack
+
+- Python
+- FastAPI
+- Pandas
+- Requests
+- Kraken public market data

submissions/mcp-hackathon/marketmind-ai/source/app/candles.py

@@ -0,0 +1,86 @@
+import requests
+import pandas as pd
+
+KRAKEN_URL = "https://api.kraken.com/0/public/OHLC"
+
+
+def get_candles(symbol: str, interval: str = "1h", limit: int = 250):
+    symbol = symbol.upper()
+
+    if not symbol.endswith("USDT"):
+        raise ValueError("Only USDT pairs are supported")
+
+    base = symbol[:-4]
+
+    # Kraken uses XBT instead of BTC
+    if base == "BTC":
+        pair = "XBTUSD"
+    else:
+        pair = f"{base}USD"
+
+    response = requests.get(
+        KRAKEN_URL,
+        params={
+            "pair": pair,
+            "interval": 60
+        },
+        timeout=15
+    )
+
+    response.raise_for_status()
+
+    result = response.json()
+
+    if result.get("error"):
+        raise RuntimeError(
+            "Kraken market data error: " + str(result["error"])
+        )
+
+    data = result["result"]
+
+    pair_key = [key for key in data.keys() if key != "last"][0]
+    candles = data[pair_key]
+
+    if not candles:
+        raise RuntimeError("No candle data returned")
+
+    df = pd.DataFrame(
+        candles,
+        columns=[
+            "open_time",
+            "open",
+            "high",
+            "low",
+            "close",
+            "vwap",
+            "volume",
+            "count"
+        ]
+    )
+
+    df["open_time"] = pd.to_datetime(
+        df["open_time"],
+        unit="s"
+    )
+
+    for column in ["open", "high", "low", "close", "volume"]:
+        df[column] = pd.to_numeric(
+            df[column],
+            errors="coerce"
+        )
+
+    df = df[
+        [
+            "open_time",
+            "open",
+            "high",
+            "low",
+            "close",
+            "volume"
+        ]
+    ]
+
+    df = df.dropna()
+    df = df.tail(limit).reset_index(drop=True)
+
+    return df

submissions/mcp-hackathon/marketmind-ai/source/app/indicators.py

@@ -0,0 +1,28 @@
+import pandas as pd
+
+
+def calculate_indicators(df):
+    df = df.copy()
+
+    df["ema20"] = df["close"].ewm(span=20, adjust=False).mean()
+    df["ema50"] = df["close"].ewm(span=50, adjust=False).mean()
+    df["ema200"] = df["close"].ewm(span=200, adjust=False).mean()
+
+    delta = df["close"].diff()
+
+    gain = delta.clip(lower=0)
+    loss = -delta.clip(upper=0)
+
+    avg_gain = gain.rolling(14).mean()
+    avg_loss = loss.rolling(14).mean()
+
+    rs = avg_gain / avg_loss.replace(0, pd.NA)
+    df["rsi14"] = 100 - (100 / (1 + rs))
+
+    ema12 = df["close"].ewm(span=12, adjust=False).mean()
+    ema26 = df["close"].ewm(span=26, adjust=False).mean()
+
+    df["macd"] = ema12 - ema26
+    df["macd_signal"] = df["macd"].ewm(span=9, adjust=False).mean()
+
+    return df

submissions/mcp-hackathon/marketmind-ai/source/app/main.py

@@ -0,0 +1,72 @@
+import os
+
+from fastapi import FastAPI
+
+from candles import get_candles
+from indicators import calculate_indicators
+from scoring import calculate_score
+
+
+app = FastAPI(
+    title="MarketMind AI",
+    description="Agent-callable crypto market intelligence API",
+    version="0.3.0"
+)
+
+
+PROJECT_SLUG = "marketmind-ai"
+
+
+@app.get("/")
+def root():
+    return {
+        "name": "MarketMind AI",
+        "status": "online",
+        "version": "0.3.0"
+    }
+
+
+@app.get("/health")
+def health():
+    commit = os.getenv("RENDER_GIT_COMMIT", "development")
+
+    return {
+        "status": "ok",
+        "commit": commit
+    }
+
+
+@app.get("/.well-known/xagent-verification.json")
+def xagent_verification():
+    commit = os.getenv("RENDER_GIT_COMMIT", "development")
+
+    return {
+        "schemaVersion": 1,
+        "slug": PROJECT_SLUG,
+        "commit": commit
+    }
+
+
+@app.get("/v1/market-intelligence/{symbol}")
+def market_intelligence(symbol: str):
+    df = get_candles(symbol)
+
+    df = calculate_indicators(df)
+
+    latest = df.iloc[-1]
+
+    score = calculate_score(latest)
+
+    return {
+        "symbol": symbol.upper(),
+        "price": float(latest["close"]),
+        "indicators": {
+            "ema20": float(latest["ema20"]),
+            "ema50": float(latest["ema50"]),
+            "ema200": float(latest["ema200"]),
+            "rsi14": float(latest["rsi14"]),
+            "macd": float(latest["macd"]),
+            "macdSignal": float(latest["macd_signal"])
+        },
+        "marketAssessment": score
+    }

submissions/mcp-hackathon/marketmind-ai/source/app/market_data.py

@@ -0,0 +1,26 @@
+import requests
+
+BINANCE_URL = "https://fapi.binance.com/fapi/v1/ticker/24hr"
+
+
+def get_market_data(symbol: str):
+    symbol = symbol.upper()
+
+    response = requests.get(
+        BINANCE_URL,
+        params={"symbol": symbol},
+        timeout=10
+    )
+
+    response.raise_for_status()
+    data = response.json()
+
+    return {
+        "symbol": data["symbol"],
+        "price": float(data["lastPrice"]),
+        "priceChange24h": float(data["priceChangePercent"]),
+        "volume24h": float(data["volume"]),
+        "quoteVolume24h": float(data["quoteVolume"]),
+        "high24h": float(data["highPrice"]),
+        "low24h": float(data["lowPrice"])
+    }

submissions/mcp-hackathon/marketmind-ai/source/app/scoring.py

@@ -0,0 +1,41 @@
+def calculate_score(row):
+    score = 50
+
+    if row["close"] > row["ema20"]:
+        score += 10
+    else:
+        score -= 10
+
+    if row["ema20"] > row["ema50"]:
+        score += 10
+    else:
+        score -= 10
+
+    if row["ema50"] > row["ema200"]:
+        score += 10
+    else:
+        score -= 10
+
+    if row["rsi14"] >= 55:
+        score += 10
+    elif row["rsi14"] < 45:
+        score -= 10
+
+    if row["macd"] > row["macd_signal"]:
+        score += 10
+    else:
+        score -= 10
+
+    score = max(0, min(100, score))
+
+    if score >= 65:
+        regime = "bullish"
+    elif score <= 35:
+        regime = "bearish"
+    else:
+        regime = "neutral"
+
+    return {
+        "score": score,
+        "regime": regime
+    }

submissions/mcp-hackathon/marketmind-ai/source/requirements.txt

@@ -0,0 +1,4 @@
+fastapi
+uvicorn
+pandas
+requests

submissions/mcp-hackathon/marketmind-ai/source/start.txt

@@ -0,0 +1 @@
+uvicorn main:app --host 0.0.0.0 --port 8000