Example
Take a ranked shortlist from the weekly benchmark surface into launch-pack generation and execution planning.
This example uses the weekly benchmark as the public shortlist layer, then hands selected papers into the same action chain that powers launch-pack generation and downstream execution.
Workflow
Proof Surface Example
Public prediction ledger with backtest calibration, frozen score decomposition, and reasoning chains for every paper.
Open proof surfacecurl
curl -X POST "https://sciencetostartup.com/api/mcp" \
-H "Authorization: Bearer s2s_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "workspace_run_action",
"arguments": {"workspace_id":"<workspace-id>","run_kind":"launch_pack"}
}
}'Python
import requests
payload = {
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "workspace_run_action",
"arguments": {"workspace_id":"<workspace-id>","run_kind":"launch_pack"},
},
}
response = requests.post(
"https://sciencetostartup.com/api/mcp",
headers={
"Authorization": "Bearer s2s_YOUR_KEY",
"Content-Type": "application/json",
},
json=payload,
timeout=30,
)TypeScript
const response = await fetch("https://sciencetostartup.com/api/mcp", {
method: "POST",
headers: {
Authorization: "Bearer s2s_YOUR_KEY",
"Content-Type": "application/json",
},
body: JSON.stringify({
jsonrpc: "2.0",
id: 1,
method: "tools/call",
params: {
name: "workspace_run_action",
arguments: {"workspace_id":"<workspace-id>","run_kind":"launch_pack"},
},
}),
});