Skip to main content

​
Overview

Set stream: true in your request to receive a real-time Server-Sent Events stream instead of waiting for the complete JSON response. This is useful for showing progress to users as the agent works.

​
Enabling Streaming

Add stream: true to your request body: 
const response = await fetch('https://api.stewrd.dev/v1/agent', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer sk-stw_your_key',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    message: 'Write a detailed analysis of renewable energy trends',
    stream: true,
  }),
})
The response will have Content-Type: text/event-stream.

​
Event Types

The stream emits the following event types:
EventDescription
tokenA chunk of the agent’s response text
tool_startThe agent is starting to use a capability (e.g., research, code)
tool_endThe agent finished using a capability
doneThe stream is complete — includes final response data
errorAn error occurred during processing

​
Event Format

Each event follows the SSE spec: 
event: token
data: {"content": "The renewable"}

event: token
data: {"content": " energy sector"}

event: tool_start
data: {"tool": "research", "message": "Searching for latest data..."}

event: tool_end
data: {"tool": "research"}

event: token
data: {"content": "According to recent findings..."}

event: done
data: {"message": "The renewable energy sector...", "files": [], "usage": {"requests_used": 5, "requests_limit": 5000, "tokens_used": 847}, "duration_ms": 4200}

​
Consuming the Stream

const response = await fetch('https://api.stewrd.dev/v1/agent', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer sk-stw_your_key',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    message: 'Explain quantum computing',
    stream: true,
  }),
})

const reader = response.body!.getReader()
const decoder = new TextDecoder()

while (true) {
  const { done, value } = await reader.read()
  if (done) break

  const chunk = decoder.decode(value)
  const lines = chunk.split('\n')

  for (const line of lines) {
    if (line.startsWith('event: ')) {
      const event = line.slice(7)
      // Handle event type
    }
    if (line.startsWith('data: ')) {
      const data = JSON.parse(line.slice(6))

      if (data.content) {
        process.stdout.write(data.content)
      }
    }
  }
}

​
Response Headers

Streaming responses include these headers:
HeaderDescription
Content-Typetext/event-stream
Cache-Controlno-cache
Connectionkeep-alive
X-Request-IdUnique request identifier
X-RateLimit-LimitMonthly request limit
X-RateLimit-RemainingRequests remaining
X-RateLimit-ResetUnix timestamp for limit reset
Usage is tracked the same way for streaming and non-streaming requests. Each request counts as one against your quota regardless of stream duration.