Getting Started

From signup to fetching news in under 5 minutes.

Quick Start

1Get your API key

2Make your first search

Search for articles

curl -X POST https://api.orbis.builder/api/v1/search \
  -H "X-API-Key: ob_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"query": "Novo Nordisk", "country": "dk", "limit": 5}'

3Create a monitoring agent

Agents continuously collect articles matching your keywords. No polling needed.

Create an agent

curl -X POST https://api.orbis.builder/api/v1/agents \
  -H "X-API-Key: ob_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Novo Nordisk Monitor",
    "keywords": [
      {"value": "Novo Nordisk", "strict": true},
      {"value": "Ozempic", "strict": true}
    ],
    "country": "dk"
  }'

4Fetch collected articles

Get agent articles

curl "https://api.orbis.builder/api/v1/agents/4521/articles?limit=10&since=2026-08-01" \
  -H "X-API-Key: ob_live_your_key_here"

Use Case: CRM News Feed

Attach live news to your CRM contacts. When a client company appears in the news, show it on their profile.

How it works

Create one agent per CRM contact with their company name as a strict keyword
Poll /agents/{id}/articles daily (or use the since parameter)
Display articles on the contact profile with source, date, and thread coverage
Use /articles/{id}/thread to show cross-site coverage when a story is big

crm_sync.py

# Python example: daily sync for a CRM contact
import requests

API_KEY = "ob_live_..."
AGENT_ID = 4521
LAST_SYNC = "2026-08-08T00:00:00Z"

resp = requests.get(
    f"https://api.orbis.builder/api/v1/agents/{AGENT_ID}/articles",
    headers={"X-API-Key": API_KEY},
    params={"since": LAST_SYNC, "limit": 50}
)

for article in resp.json()["articles"]:
    print(f"  {article['title']} ({article['source']})")
    # -> Save to CRM database

Use Case: AI Research Agent

Give your autonomous agent real-time news awareness. Use semantic search for natural language and timeline for trends.

How it works

Use the MCP server so your AI agent can call tools directly
The agent uses search_news with semantic mode for research questions
Use get_timeline to understand how a topic developed
Use get_thread to see how different sources cover the same story

AI agent tool calls

# The agent can ask natural questions:
search_news(
    query="What's happening with Danish climate policy?",
    mode="semantic",
    from_date="2026-08-01"
)

# Track coverage over time:
get_timeline(
    query="green transition Denmark",
    granularity="week"
)

# Deep-dive into a specific story:
get_thread(thread_id=7823)

Use Case: Competitor Monitoring

Track competitors across all Danish news sources. Get notified when they appear, and see which stories get the most coverage.

competitor_monitoring.py

# Create agents for each competitor
competitors = ["Danske Bank", "Nordea", "Jyske Bank", "Saxo Bank"]

for name in competitors:
    requests.post(
        "https://api.orbis.builder/api/v1/agents",
        headers={"X-API-Key": API_KEY, "Content-Type": "application/json"},
        json={
            "name": f"{name} Monitor",
            "keywords": [{"value": name, "strict": True}],
            "country": "dk"
        }
    )

# Daily: check each agent's new articles
# Weekly: compare timeline coverage between competitors

Tips

Strict vs Loose:Use strict: true for company names and exact phrases. Use strict: false for topic keywords where word order doesn't matter.
Pagination:All list endpoints support limit and offset. Check has_more to know if there are more pages.
Date filters:Always set from_date to avoid scanning the entire archive. The default is 90 days, but narrower ranges are faster.
Thread coverage:The thread_id field connects articles about the same story. Use /articles/{id}/thread to see all coverage.