Skip to main content

​
Examples

Practical examples to help you integrate the iNwealth API into your application.

​
Simple question

A single-turn question about French tax law: 
curl -X POST https://api.inwealth.fr/api/agent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d '{
    "messages": [
      {"role": "user", "content": "What are the tax brackets in France for 2025?"}
    ],
    "session_id": "demo-001",
    "residence": "fr",
    "language": "en"
  }'
For simple factual questions, leave effort and extended_thinking at their defaults to minimize token usage.

​
Multi-turn conversation

The API is stateless — send the full conversation history each time. Reuse the same session_id to benefit from prompt caching. 
# Turn 1
curl -X POST https://api.inwealth.fr/api/agent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d '{
    "messages": [
      {"role": "user", "content": "What is the PER in France?"}
    ],
    "session_id": "demo-multi-001",
    "residence": "fr",
    "language": "en"
  }'

# Turn 2 — include the full history
curl -X POST https://api.inwealth.fr/api/agent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d '{
    "messages": [
      {"role": "user", "content": "What is the PER in France?"},
      {"role": "assistant", "content": "The PER (Plan d'\''Epargne Retraite) is a retirement savings plan introduced in 2019..."},
      {"role": "user", "content": "What are the tax deduction limits for 2025?"}
    ],
    "session_id": "demo-multi-001",
    "residence": "fr",
    "language": "en"
  }'
In the assistant message, include the full text that the agent returned (all concatenated delta events). Do not include metadata.

​
Cross-border scenario

For complex cross-border questions, enable extended_thinking and use a higher effort level: 
curl -X POST https://api.inwealth.fr/api/agent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d '{
    "messages": [
      {"role": "user", "content": "I am a French tax resident with a Swiss employer. How is my salary taxed? What about the double taxation treaty?"}
    ],
    "session_id": "demo-crossborder-001",
    "residence": "fr",
    "language": "en",
    "extended_thinking": true,
    "effort": "high"
  }'

​
Multilingual support

The agent responds in the requested language, regardless of the user’s question language or residence: 
# Question in English, response in German, about Swiss taxation
curl -X POST https://api.inwealth.fr/api/agent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d '{
    "messages": [
      {"role": "user", "content": "How does the Swiss wealth tax work?"}
    ],
    "session_id": "demo-lang-001",
    "residence": "ch",
    "language": "de"
  }'

​
Token-optimized integration

Best practices to minimize token consumption in production: 
{
  "messages": [
    {"role": "user", "content": "Your question here"}
  ],
  "session_id": "user-123-conversation-456",
  "residence": "fr",
  "effort": "low"
}
Key optimizations:
  1. Reuse session_id — keep the same ID across turns in a conversation. Prompt caching can reduce token costs by up to 90%.
  2. Set effort explicitly — use "low" or "medium" for straightforward questions. Reserve "high" and "max" for complex scenarios.
  3. Trim long histories — for conversations beyond 10 turns, consider sending only the last 4-6 turns to keep input tokens low.
  4. Use extended_thinking selectively — only for questions involving multiple jurisdictions or complex tax planning.