Monitoring · Python
Polymarket Whale Watcher
AI-powered whale-trade surveillance with autonomous research and information-asymmetry scoring.
View repositoryGordon boundary
Gordon governs the project's paid agent actions. The repository keeps its own strategy, runtime, data ingestion, and execution logic.
How it works
The watcher polls Polymarket for large trades. When a whale position is detected it fires a Claude agent that calls Exa through Gordon to research whether the move is informed. The watcher passes the whale trade ID as session_id, letting Gordon group the paid calls and report what each investigation cost.
Integration recipe
How Gordon fits
Research spend
SupportedUse Gordon MCP to govern paid Exa enrichment triggered by detected whale trades.
Alert audit
SupportedAttach a session ID per investigation so Gordon groups every paid call back to the whale trade that triggered it.
Integration code
import anthropic
client = anthropic.Anthropic()
def investigate_whale_trade(trade: dict, gordon_key: str):
"""Fire a research agent for a detected whale trade."""
session_id = f"whale-{trade['id']}"
response = client.beta.messages.create(
model="claude-opus-4-5",
max_tokens=2048,
mcp_servers=[{
"type": "url",
"url": "https://api.withgordon.ai/mcp",
"name": "gordon",
"authorization_token": gordon_key,
}],
tools=[{"type": "mcp_toolset", "mcp_server_name": "gordon"}],
betas=["mcp-client-2025-11-20"],
messages=[{
"role": "user",
"content": (
f"A whale just bought ${trade['amount']:,} of YES on: {trade['market']}\n"
f"Search for recent news that might explain this move. "
f"Use session_id='{session_id}' on every gordon_call_service call."
),
}],
)
return responseConnection path
Create an agent
Give this project its own Gordon identity, wallet, and API key.
Add MCP or SDK
Use Gordon MCP for hosted agents or gordon.fetch for application code.
Set policy
Allow only required services, then cap calls and daily spend.