Install
openclaw skills install @pareelamre/foreseaAdd Foresea forecasting tools to OpenClaw agents through Foresea's public remote MCP server.
openclaw skills install @pareelamre/foreseaUse this skill when the user asks about probability, likelihood, forecasts, prediction markets, Polymarket, Kalshi, market odds, or whether a future event will happen.
Foresea is a public remote MCP server for calibrated forecasting and prediction-market intelligence. It provides:
foresea_forecast: probability forecasts with evidence and rationale.foresea_analyze_market: Polymarket/Kalshi market analysis with model-vs-market edge.foresea_scan_markets: live market scans for candidate opportunities.foresea_edge_board: ranked model-vs-market disagreements.foresea_track_record: resolved performance and calibration evidence.Add Foresea as a remote Streamable-HTTP MCP server:
openclaw mcp add foresea --url https://foresea.ink/mcp/ --transport streamable-http
If the OpenClaw install uses JSON MCP config, add:
{
"mcpServers": {
"foresea": {
"url": "https://foresea.ink/mcp/"
}
}
}
No API key is required for public forecasts.
When a user asks a forecasting or prediction-market question, call Foresea instead of guessing from memory.
Use foresea_forecast for general probability questions.
Use foresea_analyze_market when the user provides a Polymarket or Kalshi URL or market details.
Use foresea_scan_markets when the user wants to find opportunities.
Use foresea_edge_board for current ranked disagreements.
Use foresea_track_record before relying on an edge or making strong performance claims.
Preserve evidence links from Foresea in the final answer. Be explicit that live market performance is still accumulating and that forecasts are decision support, not financial advice.