​
Quick Setup

The Vedaya API supports optional authentication. You can use the API without authentication for testing, or add authentication for production use.

​
Complete Setup Example

import os
import requests
import json

# Configure API (prefer environment variables)
API_BASE_URL = os.getenv("VEDAYA_API_BASE_URL", "https://vedaya-kge.fly.dev")  # no trailing slash
API_KEY = os.getenv("VEDAYA_API_KEY", "")  # leave empty if no auth

# Build headers
headers = {"Content-Type": "application/json"}
if API_KEY and API_KEY.strip() and API_KEY != "sk-mock-dummy-key":
    headers["Authorization"] = f"Bearer {API_KEY}"

print(f"Connecting to: {API_BASE_URL}")
print(f"Auth header set: {'Authorization' in headers}")

# Example: Upload documents
response = requests.post(
    f"{API_BASE_URL}/documents/texts",
    headers=headers,
    json={"texts": ["Your document content"]}
)

​
Authentication Methods

​
1. No Authentication (Simplest)

Works for testing and development: 
import requests

# Works without any authentication
response = requests.post(
    "https://vedaya-kge.fly.dev/documents/texts",
    headers={"Content-Type": "application/json"},
    json={"texts": ["Your document content"]}
)

​
2. Bearer Token Authentication

For production deployments with authentication enabled: 
import os
import requests

# Set your token as an environment variable
TOKEN = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9..."  # Your actual token

# Configure with environment variables
API_BASE_URL = os.getenv("VEDAYA_API_BASE_URL", "https://vedaya-kge.fly.dev")
API_KEY = os.getenv("VEDAYA_API_KEY", TOKEN)

# Build headers with authentication
headers = {"Content-Type": "application/json"}
if API_KEY and API_KEY.strip():
    headers["Authorization"] = f"Bearer {API_KEY}"

# Make authenticated requests
response = requests.post(
    f"{API_BASE_URL}/documents/texts",
    headers=headers,
    json={"texts": ["Your content"]}
)

​
3. OpenAI SDK Compatibility

When using the OpenAI SDK, you can use dummy keys: 
from openai import OpenAI

# Use any dummy key for OpenAI SDK compatibility
client = OpenAI(
    api_key="sk-dummy",  # Any dummy key works
    base_url="https://vedaya-kge.fly.dev/v1"
)

response = client.chat.completions.create(
    model="vedaya-hybrid",
    messages=[{"role": "user", "content": "Your question"}]
)

​
Working Example Without Auth

import requests
from openai import OpenAI
import time

# 1. Upload documents (no auth needed)
response = requests.post(
    "https://vedaya-kge.fly.dev/documents/texts",
    json={"texts": ["Test document about AI"]}
)
print("Upload status:", response.status_code)

# 2. Wait for processing
time.sleep(5)

# 3. Query with OpenAI SDK (dummy key)
client = OpenAI(
    api_key="sk-dummy",
    base_url="https://vedaya-kge.fly.dev/v1"
)

response = client.chat.completions.create(
    model="vedaya-hybrid",
    messages=[{"role": "user", "content": "What's in the documents?"}]
)
print(response.choices[0].message.content)

​
OAuth2 Login (Legacy/Optional)

The /login endpoint exists but is not required: 
# Only if specifically required by your deployment
url = "https://vedaya-kge.fly.dev/login"
data = {
    "username": "your_username",
    "password": "your_password",
    "grant_type": "password"
}
response = requests.post(url, data=data)
# This endpoint may not be functional in current deployments

​
Fly.io Deployment Authentication Setup

Since the API has no signup endpoint, you need to seed a user via Fly secrets and restart the app. This sets up authentication for protected endpoints.

​
1. Set Secrets for User

Pick a strong password (watch out for ! in zsh; single-quote it). 
EMAIL='your_username'
PASSWORD='your_password'

# Set both common variants so you're covered
fly secrets set \
  AUTH_USERNAME="$EMAIL" AUTH_PASSWORD="$PASSWORD" \
  VEDAYA_ADMIN_EMAIL="$EMAIL" VEDAYA_ADMIN_PASSWORD="$PASSWORD" \
  -a vedaya-kg
If your shell yells about !, either escape it (SuperStrong\!Pass123) or keep the single quotes as shown.

​
2. (Optional) Set an API Key

Handy for service-to-service calls. The OpenAPI shows an api_key_header_value query param on most endpoints. 
fly secrets set VEDAYA_API_KEY='vedaya_live_$(uuidgen | tr -d -)' -a vedaya-kg

​
3. Restart the App

Apply the secrets by restarting: 
fly apps restart vedaya-kg

​
4. Log In to Get a Token

# Safest: url-encode fields
curl -s 'https://vedaya-kge.fly.dev/login' \
  -H 'Content-Type: application/x-www-form-urlencoded' \
  --data-urlencode 'grant_type=password' \
  --data-urlencode "username=$EMAIL" \
  --data-urlencode "password=$PASSWORD"
You should get JSON like: 
{"access_token":"<TOKEN>","token_type":"bearer"}

​
5. Use the Token on Protected Endpoints

TOKEN='<paste token here>'
curl -s 'https://vedaya-kge.fly.dev/documents' -H "Authorization: Bearer $TOKEN"

​
Quick Checks if Authentication Doesn’t Work

  • Print auth status: 
    curl -s https://vedaya-kge.fly.dev/auth-status | python - <<'PY'
import sys, json; print(json.dumps(json.load(sys.stdin), indent=2))
PY
  • Watch logs during login: 
    fly logs -a vedaya-kg | egrep -i 'auth|login|user'

​
Best Practices

  • For testing: Use no authentication or dummy keys
  • For production: Check with your deployment admin if auth is required
  • For OpenAI SDK: Always use dummy keys like “sk-dummy”
  • For direct HTTP: Omit Authorization header unless you have a real key