Developer Portal v2.0

Build on SpaceTime OS

The infrastructure security platform with 108+ data sources, real-time threat intelligence, and autonomous response — now with a full REST API, Python & TypeScript SDKs, and a 27-tool MCP server for AI-native agent integration.

Schedule a Demo View on GitHub

API Endpoints

MCP Tools

108+

Data Connectors

15ms

P99 Latency

99.97%

Uptime SLA

Quick Start

From zero to your first API call in under 3 minutes.

Get Access

Enterprise license — contact sales

get-api-key.sh

curl -X POST https://api.usesentinel.io/v1/auth/keys \
  -H "Content-Type: application/json" \
  -d '{ "email": "you@org.gov", "tier": "free" }'

# Response:
{
  "api_key": "sk-spacetime-xxxx...",
  "tier": "free",
  "rate_limit": "10/min"
}

Create Free Account

Install SDK

Python or TypeScript

install.sh

# Python (coming soon)
pip install spacetime-os

# TypeScript / Node
npm install @spacetime-os/edge

# Verify installation
python -c "import spacetime_os; print(spacetime_os.__version__)"
# → 2.1.0

Requires Python 3.9+ or Node 18+

Make First Call

Threat score in seconds

first_call.py

import spacetime_os as st

client = st.Client(api_key="sk-spacetime-...")

# Score a threat target
result = client.threat_score(
    target="185.220.101.45",
    target_type="ip",
    context="maritime"
)

print(result.composite_score)  # → 84
print(result.severity)         # → CRITICAL
print(result.matched_signatures)
# → ['APT40', 'VOLT-TYPHOON']

API Reference

10 production-ready endpoints. Click any card to expand the full spec.

Base URL:api.usesentinel.io/v1

Version:v2.0

27 Tools AvailableAgent-Native

MCP Server — Agent-Native Integration

SpaceTime OS ships a 27-tool MCP server that lets any AI agent — Claude, GPT, Gemini, or your custom agent — interact with the platform natively. No REST wrappers. No prompt engineering for API calls. Agents call tools directly with structured inputs and get structured outputs back.

1. Install the MCP Server

install-mcp.sh

pip install @spacetime-os/mcp-server

# Or run directly without install
uvx spacetime-os-mcp

2. Add to Claude Desktop config

claude_desktop_config.json

{
  "mcpServers": {
    "spacetime-os": {
      "command": "python",
      "args": ["-m", "spacetime_os.mcp_server"],
      "env": {
        "SPACETIME_API_KEY": "sk-spacetime-..."
      }
    }
  }
}

3. Or use with any MCP-compatible agent

mcp_client_example.py

from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client

server_params = StdioServerParameters(
    command="python",
    args=["-m", "spacetime_os.mcp_server"],
    env={"SPACETIME_API_KEY": "sk-spacetime-..."}
)

async with stdio_client(server_params) as (r, w):
    async with ClientSession(r, w) as session:
        # List all 27 tools
        tools = await session.list_tools()

        # Call a tool directly
        result = await session.call_tool(
            "get_threat_score",
            {"target": "185.220.101.45", "target_type": "ip"}
        )

Available MCP Tools

27 / 27

query_threatsSearch threats by type, location, severity

get_incidentsList and filter active incidents

create_incidentOpen a new incident from agent context

search_vesselsMaritime vessel search and profile

get_cable_riskReal-time cable proximity risk

get_threat_score7-factor composite threat scoring

get_correlationCross-domain event correlation

dispatch_droneTrigger autonomous drone response

query_data_lakeSQL query over 108+ data connectors

get_connectorsList all available data connectors

provision_free_accountSelf-serve tenant provisioning

get_space_weatherSolar/Kp index with GIC risk

get_internet_healthBGP, RPKI, IODA, Cloudflare status

get_vessel_profileZeroMemory behavioral anomaly profile

get_geopolitical_riskACLED + OFAC + CERT overlay

list_cable_systemsFull global submarine cable registry

get_ais_trackHistorical AIS vessel tracks

get_dark_vesselsCurrently dark / loitering vessels

query_threat_feedCISA KEV + abuse.ch IOC lookup

get_environmental_riskSeismic, tsunami, weather hazards

search_eventsFull-text event search with filters

get_bgp_alertsBGP hijacks and route leaks

get_sensor_dataRaw DAS/fiber optic sensor telemetry

add_watchlistAdd vessels/cables/IPs to watchlist

get_watchlist_alertsAlerts for watchlisted entities

run_playbookExecute a named response playbook

get_api_usageCheck your token quota and usage

Authentication

Bearer token auth on all endpoints. Keys are scoped per tenant with optional IP allowlists.

API Key Format

auth-header.sh

# Include in every request header:
Authorization: Bearer sk-spacetime-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

# Example with curl:
curl https://api.usesentinel.io/v1/threat-score \
  -H "Authorization: Bearer sk-spacetime-..." \
  -H "Content-Type: application/json"

Environment Variable (recommended)

.env + usage

# .env
SPACETIME_API_KEY=sk-spacetime-...

# Python — auto-discovered from env
import spacetime_os as st
client = st.Client()  # no key arg needed

# TypeScript — auto-discovered from env
import { SpaceTimeClient } from '@spacetime-os/edge'
const client = new SpaceTimeClient()

Error Codes

CODE	MEANING	ACTION
`401`	Invalid or missing API key	Check Authorization header
`403`	Endpoint not in your tier	Upgrade plan
`429`	Rate limit exceeded	Retry after X-RateLimit-Reset
`422`	Invalid request parameters	Check parameter schema
`500`	Internal server error	Retry with backoff

Rate Limits & Plans

See enterprise pricing at /pricing

ENTERPRISE

EnterpriseFrom $75k/yr

Rate limitCustom

Monthly callsUnlimited

Response headers on every call:

X-RateLimit-LimitYour tier's per-minute limit

X-RateLimit-RemainingCalls left in current window

X-RateLimit-ResetUnix timestamp for window reset

X-Request-IdUnique request ID for support

SDKs & Integration Methods

Multiple integration paths — choose what fits your stack.

🐍

Python SDK (coming soon)

spacetime-os on PyPI

PyPI

python_sdk.py

pip install spacetime-os

import spacetime_os as st

client = st.Client()

# Async support
import asyncio

async def run():
    async with st.AsyncClient() as c:
        score = await c.threat_score("185.220.101.45")
        vessel = await c.vessel_profile("413123456")
        risk = await c.cable_risk(lat=35.67, lon=139.65)

asyncio.run(run())

Python 3.9+Async/awaitType hintsAuto-retry

📘

TypeScript SDK

@spacetime-os/edge on npm

npm

ts_sdk.ts

npm install @spacetime-os/edge

import { SpaceTimeClient } from '@spacetime-os/edge'

const client = new SpaceTimeClient({
  apiKey: process.env.SPACETIME_API_KEY
})

// Fully typed responses
const score = await client.threatScore({
  target: '185.220.101.45',
  targetType: 'ip',
  context: 'maritime'
})

console.log(score.compositScore)  // number
console.log(score.severity)       // 'LOW' | 'MEDIUM' | 'HIGH' | 'CRITICAL'

TypeScript-firstNode 18+Browser supportTree-shakeable

MCP Server

Model Context Protocol · 27 tools

NEW

mcp-setup.sh

pip install @spacetime-os/mcp-server

# In Claude Desktop config:
{
  "mcpServers": {
    "spacetime-os": {
      "command": "python",
      "args": ["-m", "spacetime_os.mcp_server"],
      "env": { "SPACETIME_API_KEY": "sk-..." }
    }
  }
}

# Claude can now natively call:
# "Score this IP for threat level: 185.220.101.45"
# "Find dark vessels near cable APCN-2"
# "Check internet health for AS4837"

Claude DesktopCustom agentsGPT plugins27 tools

REST API

Direct HTTP — any language

Docs

rest-api.sh

# Works with curl, fetch, httpx, axios, Go net/http, etc.
BASE="https://api.usesentinel.io/v1"
KEY="sk-spacetime-..."

# Bash / curl
curl "$BASE/threat-score?target=185.220.101.45&target_type=ip" \
  -H "Authorization: Bearer $KEY" | jq .

# JavaScript fetch
const res = await fetch(`${BASE}/vessel-profile?mmsi=413123456`, {
  headers: { Authorization: `Bearer ${KEY}` }
})
const data = await res.json()

# Go
req, _ := http.NewRequest("GET", base+"/cable-risk", nil)
req.Header.Set("Authorization", "Bearer "+key)

OpenAPI 3.1JSON responsesHTTPS onlyCORS enabled