Spaces:
Build error
Build error
File size: 43,104 Bytes
011960a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 |
import os
import logging
from typing import Dict, List, Optional, Union, Any, Tuple
import pandas as pd
from datetime import datetime, timedelta
import io
import base64
from crewai import Agent, Task, Crew, Process
from langchain.tools import BaseTool
from langchain.chat_models import ChatOpenAI
from modules.api_client import ArbiscanClient, GeminiClient
from modules.data_processor import DataProcessor
from modules.crew_tools import (
ArbiscanGetTokenTransfersTool,
ArbiscanGetNormalTransactionsTool,
ArbiscanGetInternalTransactionsTool,
ArbiscanFetchWhaleTransactionsTool,
GeminiGetCurrentPriceTool,
GeminiGetHistoricalPricesTool,
DataProcessorIdentifyPatternsTool,
DataProcessorDetectAnomalousTransactionsTool,
set_global_clients
)
class WhaleAnalysisCrewSystem:
"""
CrewAI system for analyzing whale wallet activity and detecting market manipulation
"""
def __init__(self, arbiscan_client: ArbiscanClient, gemini_client: GeminiClient, data_processor: DataProcessor):
self.arbiscan_client = arbiscan_client
self.gemini_client = gemini_client
self.data_processor = data_processor
# Initialize LLM
try:
from langchain.chat_models import ChatOpenAI
self.llm = ChatOpenAI(
model="gpt-4",
temperature=0.2,
api_key=os.getenv("OPENAI_API_KEY")
)
except Exception as e:
logging.warning(f"Could not initialize LLM: {str(e)}")
self.llm = None
# Use a factory method to safely create tool instances
self.setup_tools()
def setup_tools(self):
"""Setup LangChain tools for the whale analysis crew"""
try:
# Setup clients
arbiscan_client = ArbiscanClient(api_key=os.getenv("ARBISCAN_API_KEY"))
gemini_client = GeminiClient(api_key=os.getenv("GEMINI_API_KEY"))
data_processor = DataProcessor()
# Set global clients first
set_global_clients(
arbiscan_client=arbiscan_client,
gemini_client=gemini_client,
data_processor=data_processor
)
# Create tools (no need to pass clients, they'll use globals)
self.arbiscan_tools = [
self._create_tool(ArbiscanGetTokenTransfersTool),
self._create_tool(ArbiscanGetNormalTransactionsTool),
self._create_tool(ArbiscanGetInternalTransactionsTool),
self._create_tool(ArbiscanFetchWhaleTransactionsTool)
]
self.gemini_tools = [
self._create_tool(GeminiGetCurrentPriceTool),
self._create_tool(GeminiGetHistoricalPricesTool)
]
self.data_processor_tools = [
self._create_tool(DataProcessorIdentifyPatternsTool),
self._create_tool(DataProcessorDetectAnomalousTransactionsTool)
]
logging.info(f"Successfully created {len(self.arbiscan_tools + self.gemini_tools + self.data_processor_tools)} tools")
except Exception as e:
logging.error(f"Error setting up tools: {str(e)}")
raise Exception(f"Error setting up tools: {str(e)}")
def _create_tool(self, tool_class, *args, **kwargs):
"""Factory method to safely create a tool with proper error handling"""
try:
tool = tool_class(*args, **kwargs)
return tool
except Exception as e:
logging.error(f"Failed to create tool {tool_class.__name__}: {str(e)}")
raise Exception(f"Failed to create tool {tool_class.__name__}: {str(e)}")
def create_agents(self):
"""Create the agents for the crew"""
# Data Collection Agent
data_collector = Agent(
role="Blockchain Data Collector",
goal="Collect comprehensive whale transaction data from the blockchain",
backstory="""You are a blockchain analytics expert specialized in extracting and
organizing on-chain data from the Arbitrum network. You have deep knowledge of blockchain
transaction structures and can efficiently query APIs to gather relevant whale activity.""",
verbose=True,
allow_delegation=True,
tools=self.arbiscan_tools,
llm=self.llm
)
# Price Analysis Agent
price_analyst = Agent(
role="Price Impact Analyst",
goal="Analyze how whale transactions impact token prices",
backstory="""You are a quantitative market analyst with expertise in correlating
trading activity with price movements. You specialize in detecting how large trades
influence market dynamics, and can identify unusual price patterns.""",
verbose=True,
allow_delegation=True,
tools=self.gemini_tools,
llm=self.llm
)
# Pattern Detection Agent
pattern_detector = Agent(
role="Trading Pattern Detector",
goal="Identify recurring behavior patterns in whale trading activity",
backstory="""You are a data scientist specialized in time-series analysis and behavioral
pattern recognition. You excel at spotting cyclical behaviors, correlation patterns, and
anomalous trading activities across multiple addresses.""",
verbose=True,
allow_delegation=True,
tools=self.data_processor_tools,
llm=self.llm
)
# Manipulation Detector Agent
manipulation_detector = Agent(
role="Market Manipulation Investigator",
goal="Detect potential market manipulation in whale activity",
backstory="""You are a financial forensics expert who has studied market manipulation
techniques for years. You can identify pump-and-dump schemes, wash trading, spoofing,
and other deceptive practices used by whale traders to manipulate market prices.""",
verbose=True,
allow_delegation=True,
tools=self.data_processor_tools,
llm=self.llm
)
# Report Generator Agent
report_generator = Agent(
role="Insights Reporter",
goal="Create comprehensive, actionable reports on whale activity",
backstory="""You are a financial data storyteller who excels at transforming complex
blockchain data into clear, insightful narratives. You can distill technical findings
into actionable intelligence for different audiences.""",
verbose=True,
allow_delegation=True,
tools=[],
llm=self.llm
)
return {
"data_collector": data_collector,
"price_analyst": price_analyst,
"pattern_detector": pattern_detector,
"manipulation_detector": manipulation_detector,
"report_generator": report_generator
}
def track_large_transactions(self,
wallets: List[str],
start_date: datetime,
end_date: datetime,
threshold_value: float,
threshold_type: str,
token_symbol: Optional[str] = None) -> pd.DataFrame:
"""
Track large buy/sell transactions for specified wallets
Args:
wallets: List of wallet addresses to track
start_date: Start date for analysis
end_date: End date for analysis
threshold_value: Minimum value for transaction tracking
threshold_type: Type of threshold ("Token Amount" or "USD Value")
token_symbol: Symbol of token to track (only required if threshold_type is "Token Amount")
Returns:
DataFrame of large transactions
"""
agents = self.create_agents()
# Define tasks
data_collection_task = Task(
description=f"""
Collect all transactions for the following wallets: {', '.join(wallets)}
between {start_date.strftime('%Y-%m-%d')} and {end_date.strftime('%Y-%m-%d')}.
Filter for transactions {'of ' + token_symbol if token_symbol else ''} with a
{'token amount greater than ' + str(threshold_value) if threshold_type == 'Token Amount'
else 'USD value greater than $' + str(threshold_value)}.
Return the data in a well-structured format with timestamp, transaction hash,
sender, recipient, token symbol, and amount.
""",
agent=agents["data_collector"],
expected_output="""
A comprehensive dataset of all large transactions for the specified wallets,
properly filtered according to the threshold criteria.
"""
)
# Create and run the crew
crew = Crew(
agents=[agents["data_collector"]],
tasks=[data_collection_task],
verbose=2,
process=Process.sequential
)
result = crew.kickoff()
# Process the result
import json
try:
# Try to extract JSON from the result
import re
json_match = re.search(r'```json\n([\s\S]*?)\n```', result)
if json_match:
json_str = json_match.group(1)
transactions_data = json.loads(json_str)
if isinstance(transactions_data, list):
return pd.DataFrame(transactions_data)
else:
return pd.DataFrame()
else:
# Try to parse the entire result as JSON
transactions_data = json.loads(result)
if isinstance(transactions_data, list):
return pd.DataFrame(transactions_data)
else:
return pd.DataFrame()
except:
# Fallback to querying the API directly
token_address = None # Would need a mapping of symbol to address
transactions_df = self.arbiscan_client.fetch_whale_transactions(
addresses=wallets,
token_address=token_address,
min_token_amount=threshold_value if threshold_type == "Token Amount" else None,
min_usd_value=threshold_value if threshold_type == "USD Value" else None
)
return transactions_df
def identify_trading_patterns(self,
wallets: List[str],
start_date: datetime,
end_date: datetime) -> List[Dict[str, Any]]:
"""
Identify trading patterns for specified wallets
Args:
wallets: List of wallet addresses to analyze
start_date: Start date for analysis
end_date: End date for analysis
Returns:
List of identified patterns
"""
agents = self.create_agents()
# Define tasks
data_collection_task = Task(
description=f"""
Collect all transactions for the following wallets: {', '.join(wallets)}
between {start_date.strftime('%Y-%m-%d')} and {end_date.strftime('%Y-%m-%d')}.
Include all token transfers, regardless of size.
""",
agent=agents["data_collector"],
expected_output="""
A comprehensive dataset of all transactions for the specified wallets.
"""
)
pattern_analysis_task = Task(
description="""
Analyze the transaction data to identify recurring trading patterns.
Look for:
1. Cyclical buying/selling behaviors
2. Time-of-day patterns
3. Accumulation/distribution phases
4. Coordinated movements across multiple addresses
Cluster similar behaviors and describe each pattern identified.
""",
agent=agents["pattern_detector"],
expected_output="""
A detailed analysis of trading patterns with:
- Pattern name/type
- Description of behavior
- Frequency and confidence level
- Example transactions showing the pattern
""",
context=[data_collection_task]
)
# Create and run the crew
crew = Crew(
agents=[agents["data_collector"], agents["pattern_detector"]],
tasks=[data_collection_task, pattern_analysis_task],
verbose=2,
process=Process.sequential
)
result = crew.kickoff()
# Process the result
import json
try:
# Try to extract JSON from the result
import re
json_match = re.search(r'```json\n([\s\S]*?)\n```', result)
if json_match:
json_str = json_match.group(1)
patterns_data = json.loads(json_str)
# Convert the patterns to the expected format
return self._convert_patterns_to_visual_format(patterns_data)
else:
# Fallback to a simple pattern analysis
# First, get transaction data directly
all_transactions = []
for wallet in wallets:
transfers = self.arbiscan_client.fetch_all_token_transfers(
address=wallet
)
all_transactions.extend(transfers)
if not all_transactions:
return []
transactions_df = pd.DataFrame(all_transactions)
# Use data processor to identify patterns
patterns = self.data_processor.identify_patterns(transactions_df)
return patterns
except Exception as e:
print(f"Error processing patterns: {str(e)}")
return []
def _convert_patterns_to_visual_format(self, patterns_data: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
"""
Convert pattern data from agents to visual format with charts
Args:
patterns_data: Pattern data from agents
Returns:
List of patterns with visualizations
"""
visual_patterns = []
for pattern in patterns_data:
# Create chart
if 'examples' in pattern and pattern['examples']:
examples_data = []
# Check if examples is a JSON string
if isinstance(pattern['examples'], str):
try:
examples_data = pd.read_json(pattern['examples'])
except:
examples_data = pd.DataFrame()
else:
examples_data = pd.DataFrame(pattern['examples'])
# Create visualization
if not examples_data.empty:
import plotly.express as px
# Check for timestamp column
if 'Timestamp' in examples_data.columns:
time_col = 'Timestamp'
elif 'timeStamp' in examples_data.columns:
time_col = 'timeStamp'
else:
time_col = None
# Check for amount column
if 'Amount' in examples_data.columns:
amount_col = 'Amount'
elif 'tokenAmount' in examples_data.columns:
amount_col = 'tokenAmount'
elif 'value' in examples_data.columns:
amount_col = 'value'
else:
amount_col = None
if time_col and amount_col:
# Create time series chart
fig = px.line(
examples_data,
x=time_col,
y=amount_col,
title=f"Pattern: {pattern['name']}"
)
else:
fig = None
else:
fig = None
else:
fig = None
examples_data = pd.DataFrame()
# Create visual pattern object
visual_pattern = {
"name": pattern.get("name", "Unknown Pattern"),
"description": pattern.get("description", ""),
"confidence": pattern.get("confidence", 0.5),
"occurrence_count": pattern.get("occurrence_count", 0),
"chart_data": fig,
"examples": examples_data
}
visual_patterns.append(visual_pattern)
return visual_patterns
def analyze_price_impact(self,
wallets: List[str],
start_date: datetime,
end_date: datetime,
lookback_minutes: int = 5,
lookahead_minutes: int = 5) -> Dict[str, Any]:
"""
Analyze the impact of whale transactions on token prices
Args:
wallets: List of wallet addresses to analyze
start_date: Start date for analysis
end_date: End date for analysis
lookback_minutes: Minutes to look back before transactions
lookahead_minutes: Minutes to look ahead after transactions
Returns:
Dictionary with price impact analysis
"""
agents = self.create_agents()
# Define tasks
data_collection_task = Task(
description=f"""
Collect all transactions for the following wallets: {', '.join(wallets)}
between {start_date.strftime('%Y-%m-%d')} and {end_date.strftime('%Y-%m-%d')}.
Focus on large transactions that might impact price.
""",
agent=agents["data_collector"],
expected_output="""
A comprehensive dataset of all significant transactions for the specified wallets.
"""
)
price_impact_task = Task(
description=f"""
Analyze the price impact of the whale transactions.
For each transaction:
1. Fetch price data for {lookback_minutes} minutes before and {lookahead_minutes} minutes after the transaction
2. Calculate the percentage price change
3. Identify transactions that caused significant price moves
Summarize the overall price impact statistics and highlight notable instances.
""",
agent=agents["price_analyst"],
expected_output="""
A detailed analysis of price impacts with:
- Average price impact percentage
- Maximum price impact (positive and negative)
- Count of significant price moves
- List of transactions with their corresponding price impacts
""",
context=[data_collection_task]
)
# Create and run the crew
crew = Crew(
agents=[agents["data_collector"], agents["price_analyst"]],
tasks=[data_collection_task, price_impact_task],
verbose=2,
process=Process.sequential
)
result = crew.kickoff()
# Process the result
import json
try:
# Try to extract JSON from the result
import re
json_match = re.search(r'```json\n([\s\S]*?)\n```', result)
if json_match:
json_str = json_match.group(1)
impact_data = json.loads(json_str)
# Convert the impact data to visual format
return self._convert_impact_to_visual_format(impact_data)
else:
# Fallback to direct calculation
# First, get transaction data
all_transactions = []
for wallet in wallets:
transfers = self.arbiscan_client.fetch_all_token_transfers(
address=wallet
)
all_transactions.extend(transfers)
if not all_transactions:
return {}
transactions_df = pd.DataFrame(all_transactions)
# Calculate price impact for each transaction
price_data = {}
for idx, row in transactions_df.iterrows():
tx_hash = row.get('hash', '')
if not tx_hash:
continue
# Get symbol
symbol = row.get('tokenSymbol', '')
if not symbol:
continue
# Get timestamp
timestamp = row.get('timeStamp', 0)
if not timestamp:
continue
# Convert timestamp to datetime
if isinstance(timestamp, (int, float)):
tx_time = datetime.fromtimestamp(int(timestamp))
else:
tx_time = timestamp
# Get price impact
symbol_usd = f"{symbol}USD"
impact = self.gemini_client.get_price_impact(
symbol=symbol_usd,
transaction_time=tx_time,
lookback_minutes=lookback_minutes,
lookahead_minutes=lookahead_minutes
)
price_data[tx_hash] = impact
# Use data processor to analyze price impact
impact_analysis = self.data_processor.analyze_price_impact(
transactions_df=transactions_df,
price_data=price_data
)
return impact_analysis
except Exception as e:
print(f"Error processing price impact: {str(e)}")
return {}
def _convert_impact_to_visual_format(self, impact_data: Dict[str, Any]) -> Dict[str, Any]:
"""
Convert price impact data to visual format with charts
Args:
impact_data: Price impact data
Returns:
Dictionary with price impact analysis and visualizations
"""
# Convert transactions_with_impact to DataFrame if it's a string
if 'transactions_with_impact' in impact_data and isinstance(impact_data['transactions_with_impact'], str):
try:
transactions_df = pd.read_json(impact_data['transactions_with_impact'])
except:
transactions_df = pd.DataFrame()
elif 'transactions_with_impact' in impact_data and isinstance(impact_data['transactions_with_impact'], list):
transactions_df = pd.DataFrame(impact_data['transactions_with_impact'])
else:
transactions_df = pd.DataFrame()
# Create impact chart
if not transactions_df.empty and 'impact_pct' in transactions_df.columns and 'Timestamp' in transactions_df.columns:
import plotly.graph_objects as go
fig = go.Figure()
fig.add_trace(go.Scatter(
x=transactions_df['Timestamp'],
y=transactions_df['impact_pct'],
mode='markers+lines',
name='Price Impact (%)',
marker=dict(
size=10,
color=transactions_df['impact_pct'],
colorscale='RdBu',
cmin=-max(abs(transactions_df['impact_pct'])) if len(transactions_df) > 0 else -1,
cmax=max(abs(transactions_df['impact_pct'])) if len(transactions_df) > 0 else 1,
colorbar=dict(title='Impact %'),
symbol='circle'
)
))
fig.update_layout(
title='Price Impact of Whale Transactions',
xaxis_title='Timestamp',
yaxis_title='Price Impact (%)',
hovermode='closest'
)
# Add zero line
fig.add_hline(y=0, line_dash="dash", line_color="gray")
else:
fig = None
# Create visual impact analysis
visual_impact = {
'avg_impact_pct': impact_data.get('avg_impact_pct', 0),
'max_impact_pct': impact_data.get('max_impact_pct', 0),
'min_impact_pct': impact_data.get('min_impact_pct', 0),
'significant_moves_count': impact_data.get('significant_moves_count', 0),
'total_transactions': impact_data.get('total_transactions', 0),
'impact_chart': fig,
'transactions_with_impact': transactions_df
}
return visual_impact
def detect_manipulation(self,
wallets: List[str],
start_date: datetime,
end_date: datetime,
sensitivity: str = "Medium") -> List[Dict[str, Any]]:
"""
Detect potential market manipulation by whale wallets
Args:
wallets: List of wallet addresses to analyze
start_date: Start date for analysis
end_date: End date for analysis
sensitivity: Detection sensitivity ("Low", "Medium", "High")
Returns:
List of manipulation alerts
"""
agents = self.create_agents()
# Define tasks
data_collection_task = Task(
description=f"""
Collect all transactions for the following wallets: {', '.join(wallets)}
between {start_date.strftime('%Y-%m-%d')} and {end_date.strftime('%Y-%m-%d')}.
Include all token transfers and also fetch price data if available.
""",
agent=agents["data_collector"],
expected_output="""
A comprehensive dataset of all transactions for the specified wallets.
"""
)
price_impact_task = Task(
description="""
Analyze the price impact of the whale transactions.
For each significant transaction, fetch and analyze price data around the transaction time.
""",
agent=agents["price_analyst"],
expected_output="""
Price impact data for the transactions.
""",
context=[data_collection_task]
)
manipulation_detection_task = Task(
description=f"""
Detect potential market manipulation patterns in the transaction data with sensitivity level: {sensitivity}.
Look for:
1. Pump-and-Dump: Rapid buys followed by coordinated sell-offs
2. Wash Trading: Self-trading across multiple addresses
3. Spoofing: Large orders placed then canceled (if detectable)
4. Momentum Ignition: Creating sharp price moves to trigger other participants' momentum-based trading
For each potential manipulation, provide:
- Type of manipulation
- Involved addresses
- Risk level (High, Medium, Low)
- Description of the suspicious behavior
- Evidence (transactions showing the pattern)
""",
agent=agents["manipulation_detector"],
expected_output="""
A detailed list of potential manipulation incidents with supporting evidence.
""",
context=[data_collection_task, price_impact_task]
)
# Create and run the crew
crew = Crew(
agents=[
agents["data_collector"],
agents["price_analyst"],
agents["manipulation_detector"]
],
tasks=[
data_collection_task,
price_impact_task,
manipulation_detection_task
],
verbose=2,
process=Process.sequential
)
result = crew.kickoff()
# Process the result
import json
try:
# Try to extract JSON from the result
import re
json_match = re.search(r'```json\n([\s\S]*?)\n```', result)
if json_match:
json_str = json_match.group(1)
alerts_data = json.loads(json_str)
# Convert the alerts to visual format
return self._convert_alerts_to_visual_format(alerts_data)
else:
# Fallback to direct detection
# First, get transaction data
all_transactions = []
for wallet in wallets:
transfers = self.arbiscan_client.fetch_all_token_transfers(
address=wallet
)
all_transactions.extend(transfers)
if not all_transactions:
return []
transactions_df = pd.DataFrame(all_transactions)
# Calculate price impact for each transaction
price_data = {}
for idx, row in transactions_df.iterrows():
tx_hash = row.get('hash', '')
if not tx_hash:
continue
# Get symbol
symbol = row.get('tokenSymbol', '')
if not symbol:
continue
# Get timestamp
timestamp = row.get('timeStamp', 0)
if not timestamp:
continue
# Convert timestamp to datetime
if isinstance(timestamp, (int, float)):
tx_time = datetime.fromtimestamp(int(timestamp))
else:
tx_time = timestamp
# Get price impact
symbol_usd = f"{symbol}USD"
impact = self.gemini_client.get_price_impact(
symbol=symbol_usd,
transaction_time=tx_time,
lookback_minutes=5,
lookahead_minutes=5
)
price_data[tx_hash] = impact
# Detect wash trading
wash_trading_alerts = self.data_processor.detect_wash_trading(
transactions_df=transactions_df,
addresses=wallets,
sensitivity=sensitivity
)
# Detect pump and dump
pump_and_dump_alerts = self.data_processor.detect_pump_and_dump(
transactions_df=transactions_df,
price_data=price_data,
sensitivity=sensitivity
)
# Combine alerts
all_alerts = wash_trading_alerts + pump_and_dump_alerts
return all_alerts
except Exception as e:
print(f"Error detecting manipulation: {str(e)}")
return []
def _convert_alerts_to_visual_format(self, alerts_data: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
"""
Convert manipulation alerts data to visual format with charts
Args:
alerts_data: Alerts data from agents
Returns:
List of alerts with visualizations
"""
visual_alerts = []
for alert in alerts_data:
# Create chart based on alert type
if 'evidence' in alert and alert['evidence']:
evidence_data = []
# Check if evidence is a JSON string
if isinstance(alert['evidence'], str):
try:
evidence_data = pd.read_json(alert['evidence'])
except:
evidence_data = pd.DataFrame()
else:
evidence_data = pd.DataFrame(alert['evidence'])
# Create visualization based on alert type
if not evidence_data.empty:
import plotly.graph_objects as go
import plotly.express as px
# Check for timestamp column
if 'Timestamp' in evidence_data.columns:
time_col = 'Timestamp'
elif 'timeStamp' in evidence_data.columns:
time_col = 'timeStamp'
elif 'timestamp' in evidence_data.columns:
time_col = 'timestamp'
else:
time_col = None
# Different visualizations based on alert type
if alert.get('type') == 'Wash Trading' and time_col:
# Create scatter plot of wash trading
fig = px.scatter(
evidence_data,
x=time_col,
y=evidence_data.get('Amount', evidence_data.get('tokenAmount', evidence_data.get('value', 0))),
color=evidence_data.get('From', evidence_data.get('from', 'Unknown')),
title=f"Wash Trading Evidence: {alert.get('title', '')}"
)
elif alert.get('type') == 'Pump and Dump' and time_col and 'pre_price' in evidence_data.columns:
# Create price line for pump and dump
fig = go.Figure()
# Plot price line
fig.add_trace(go.Scatter(
x=evidence_data[time_col],
y=evidence_data['pre_price'],
mode='lines+markers',
name='Price Before Transaction',
line=dict(color='blue')
))
fig.add_trace(go.Scatter(
x=evidence_data[time_col],
y=evidence_data['post_price'],
mode='lines+markers',
name='Price After Transaction',
line=dict(color='red')
))
fig.update_layout(
title=f"Pump and Dump Evidence: {alert.get('title', '')}",
xaxis_title='Time',
yaxis_title='Price',
hovermode='closest'
)
elif alert.get('type') == 'Momentum Ignition' and time_col and 'impact_pct' in evidence_data.columns:
# Create impact scatter for momentum ignition
fig = px.scatter(
evidence_data,
x=time_col,
y='impact_pct',
size=abs(evidence_data['impact_pct']),
color='impact_pct',
color_continuous_scale='RdBu',
title=f"Momentum Ignition Evidence: {alert.get('title', '')}"
)
else:
# Generic timeline view
if time_col:
fig = px.timeline(
evidence_data,
x_start=time_col,
x_end=time_col,
y=evidence_data.get('From', evidence_data.get('from', 'Unknown')),
color=alert.get('risk_level', 'Medium'),
title=f"Alert Evidence: {alert.get('title', '')}"
)
else:
fig = None
else:
fig = None
else:
fig = None
evidence_data = pd.DataFrame()
# Create visual alert object
visual_alert = {
"type": alert.get("type", "Unknown"),
"addresses": alert.get("addresses", []),
"risk_level": alert.get("risk_level", "Medium"),
"description": alert.get("description", ""),
"detection_time": alert.get("detection_time", datetime.now().strftime("%Y-%m-%d %H:%M:%S")),
"title": alert.get("title", "Alert"),
"evidence": evidence_data,
"chart": fig
}
visual_alerts.append(visual_alert)
return visual_alerts
def generate_report(self,
wallets: List[str],
start_date: datetime,
end_date: datetime,
report_type: str = "Transaction Summary",
export_format: str = "PDF") -> Dict[str, Any]:
"""
Generate a report of whale activity
Args:
wallets: List of wallet addresses to include in the report
start_date: Start date for report period
end_date: End date for report period
report_type: Type of report to generate
export_format: Format for the report (CSV, PDF, PNG)
Returns:
Dictionary with report data
"""
from modules.visualizer import Visualizer
visualizer = Visualizer()
agents = self.create_agents()
# Define tasks
data_collection_task = Task(
description=f"""
Collect all transactions for the following wallets: {', '.join(wallets)}
between {start_date.strftime('%Y-%m-%d')} and {end_date.strftime('%Y-%m-%d')}.
""",
agent=agents["data_collector"],
expected_output="""
A comprehensive dataset of all transactions for the specified wallets.
"""
)
report_task = Task(
description=f"""
Generate a {report_type} report in {export_format} format.
The report should include:
1. Executive summary of wallet activity
2. Transaction analysis
3. Pattern identification (if applicable)
4. Price impact analysis (if applicable)
5. Manipulation detection (if applicable)
Organize the information clearly and provide actionable insights.
""",
agent=agents["report_generator"],
expected_output=f"""
A complete {export_format} report with all relevant analyses.
""",
context=[data_collection_task]
)
# Create and run the crew
crew = Crew(
agents=[agents["data_collector"], agents["report_generator"]],
tasks=[data_collection_task, report_task],
verbose=2,
process=Process.sequential
)
result = crew.kickoff()
# Process the result - for reports, we'll use our visualizer directly
# First, get transaction data
all_transactions = []
for wallet in wallets:
transfers = self.arbiscan_client.fetch_all_token_transfers(
address=wallet
)
all_transactions.extend(transfers)
if not all_transactions:
return {
"filename": f"no_data_report_{datetime.now().strftime('%Y%m%d_%H%M%S')}.{export_format.lower()}",
"content": ""
}
transactions_df = pd.DataFrame(all_transactions)
# Generate the report based on format
filename = f"whale_report_{datetime.now().strftime('%Y%m%d_%H%M%S')}"
if export_format == "CSV":
content = visualizer.generate_csv_report(
transactions_df=transactions_df,
report_type=report_type
)
filename += ".csv"
return {
"filename": filename,
"content": content
}
elif export_format == "PDF":
# For PDF we need to get more data
# Run pattern detection
patterns = self.identify_trading_patterns(
wallets=wallets,
start_date=start_date,
end_date=end_date
)
# Run price impact analysis
price_impact = self.analyze_price_impact(
wallets=wallets,
start_date=start_date,
end_date=end_date
)
# Run manipulation detection
alerts = self.detect_manipulation(
wallets=wallets,
start_date=start_date,
end_date=end_date
)
content = visualizer.generate_pdf_report(
transactions_df=transactions_df,
patterns=patterns,
price_impact=price_impact,
alerts=alerts,
title=f"Whale Analysis Report: {report_type}",
start_date=start_date,
end_date=end_date
)
filename += ".pdf"
return {
"filename": filename,
"content": content
}
elif export_format == "PNG":
# For PNG we'll create a chart based on report type
if report_type == "Transaction Summary":
fig = visualizer.create_transaction_timeline(transactions_df)
elif report_type == "Pattern Analysis":
fig = visualizer.create_volume_chart(transactions_df)
elif report_type == "Price Impact":
# Run price impact analysis first
price_impact = self.analyze_price_impact(
wallets=wallets,
start_date=start_date,
end_date=end_date
)
fig = price_impact.get('impact_chart', visualizer.create_transaction_timeline(transactions_df))
else: # "Manipulation Detection" or "Complete Analysis"
fig = visualizer.create_network_graph(transactions_df)
content = visualizer.generate_png_chart(fig)
filename += ".png"
return {
"filename": filename,
"content": content
}
else:
return {
"filename": f"unsupported_format_{datetime.now().strftime('%Y%m%d_%H%M%S')}.txt",
"content": "Unsupported export format requested."
}
|