Quickstart

This is the shortest path from nothing to a working call. It takes about five minutes.

1. Get an API key#

Sign in at oden-api.com/app and generate a key from the dashboard. Every account starts on the free tier: 1,000 searches a month, no credit card. Keys look like oden_live_….

Keep the key server-side. Anyone who has it can spend your quota, so put it in an environment variable rather than in your source:

export ODEN_KEY="oden_live_your_key_here"

2. Make your first call#

Send a query to POST /search. That is the whole API.

curl https://api.oden-api.com/search \
  -H "Authorization: Bearer $ODEN_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "What are the symptoms of vitamin D deficiency?"}'

The same call in Python (standard library plus requests):

import os, requests

resp = requests.post(
    "https://api.oden-api.com/search",
    headers={"Authorization": f"Bearer {os.environ['ODEN_KEY']}"},
    json={"query": "What are the symptoms of vitamin D deficiency?", "answer": True},
    timeout=30,
)
resp.raise_for_status()
data = resp.json()
print(data["results"]["answer"])
for c in data["results"]["citations"]:
    print(f"- {c['title']} ({c['url']})")

In JavaScript / Node (18+, built-in fetch):

const resp = await fetch("https://api.oden-api.com/search", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.ODEN_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ query: "What are the symptoms of vitamin D deficiency?", answer: true }),
});
if (!resp.ok) throw new Error(`ODEN ${resp.status}`);
const data = await resp.json();
console.log(data.results.answer);

In TypeScript, with the response typed:

interface OdenCitation { title: string; url: string; score: number; snippet?: string }
interface OdenResponse {
  query: string;
  trace_id: string;
  cached: boolean;
  results: { answer?: string; citations: OdenCitation[] };
}

async function search(query: string): Promise<OdenResponse> {
  const resp = await fetch("https://api.oden-api.com/search", {
    method: "POST",
    headers: {
      Authorization: `Bearer ${process.env.ODEN_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({ query }),
  });
  if (!resp.ok) throw new Error(`ODEN request failed: ${resp.status}`);
  return resp.json() as Promise<OdenResponse>;
}

3. Read the response#

You get the answer and the sources in one object:

{
  "query": "What are the symptoms of vitamin D deficiency?",
  "trace_id": "oden-mquycpae-p3thrk",
  "cached": false,
  "results": {
    "answer": "Low vitamin D can cause muscle weakness, bone pain and fatigue...",
    "citations": [
      { "title": "Vitamin D deficiency — 1177 Vårdguiden", "url": "https://www.1177.se/...", "score": 0.912 }
    ]
  }
}

• results.answer — the synthesized answer, in the same language as your query.
• results.citations — the ranked sources, highest score first.
• trace_id — include it if you ever report an issue.
• cached — true when the result came from cache and returned in tens of milliseconds.

On Windows PowerShell, curl mangles the inner quotes in the JSON body. Use Invoke-RestMethod with ConvertTo-Json, or run the curl example from WSL or Git Bash.

Next steps#

• Tune the call with the search parameters — answer, depth and include_snippets.
• Understand answers and citations in depth.
• Handle rate limits and errors before you go to production.
• Wire ODEN into a RAG pipeline or OpenAI function calling.