Code Examples

Copy-paste integrations for the languages you're most likely to use. Every snippet calls /v1/chat with two headers and a message body — same shape as Authentication and the Chat Endpoint.

Hello, agent — single call

bash

curl -X POST https://api.sentientone.ai/v1/chat \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $SENTIENTONE_KEY" \
  -H "X-Agent-Id: $SENTIENTONE_AGENT" \
  -d '{ "message": "Hello, agent!" }'

Multi-turn — continuing a conversation

bash

# 1) Start a thread
curl -X POST https://api.sentientone.ai/v1/chat \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $SENTIENTONE_KEY" \
  -H "X-Agent-Id: $SENTIENTONE_AGENT" \
  -d '{ "message": "Hi, my order ORD-1234 hasn'\''t arrived." }'

# Response includes "conversation_id": "conv-abc123"

# 2) Continue it
curl -X POST https://api.sentientone.ai/v1/chat \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $SENTIENTONE_KEY" \
  -H "X-Agent-Id: $SENTIENTONE_AGENT" \
  -d '{ "conversation_id": "conv-abc123", "message": "What carrier is it on?" }'

Streaming — token-by-token

bash

curl -N -X POST https://api.sentientone.ai/v1/chat/stream \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $SENTIENTONE_KEY" \
  -H "X-Agent-Id: $SENTIENTONE_AGENT" \
  -d '{ "message": "Stream me a haiku." }'

Best practices

Store X-Api-Key in your server environment, never in client code. Always proxy through your backend.
Persist the conversation_id per end-user session so follow-up messages keep context.
Honour Retry-After on 429 and use exponential backoff on 5xx — see Rate Limits.
Log the X-Trace-Id from every response alongside your application logs so you can pivot into Tracing in one click.
For chat UIs, use SSE instead of polling — the perceived latency is dramatically better.

Need a request with your real values?

Use the API Sandbox and click Copy as cURL / Copy as TypeScript / Copy as Python. The snippet is generated with your real headers and your actual request body.