Skip to main content
Jenkins Job Insight supports two analysis modes to accommodate different integration patterns. Choose the mode that best fits your workflow.

​
Async Mode (Default)

Async mode returns immediately with a job ID, allowing you to poll for results or receive them via webhook callback. This is ideal for long-running analyses and CI/CD integrations.

​
How It Works

  1. Submit analysis request to /analyze
  2. Receive job ID immediately (HTTP 202)
  3. Poll /results/{job_id} for status updates
  4. Or configure a callback webhook to receive results automatically

​
Example: Async with Polling

curl -X POST http://localhost:8000/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "job_name": "test-pipeline",
    "build_number": 123,
    "ai_provider": "claude",
    "ai_model": "claude-sonnet-4-20250514"
  }'
Response: 
{
  "status": "queued",
  "job_id": "550e8400-e29b-41d4-a716-446655440000",
  "message": "Analysis job queued. Poll /results/{job_id} for status.",
  "base_url": "http://localhost:8000",
  "result_url": "http://localhost:8000/results/550e8400-e29b-41d4-a716-446655440000",
  "html_report_url": "http://localhost:8000/results/550e8400-e29b-41d4-a716-446655440000.html"
}

​
Example: Async with Callback

Configure a webhook to receive results automatically when analysis completes:
cURL
curl -X POST http://localhost:8000/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "job_name": "test-pipeline",
    "build_number": 123,
    "ai_provider": "claude",
    "ai_model": "claude-sonnet-4-20250514",
    "callback_url": "https://your-service.com/webhook",
    "callback_headers": {
      "Authorization": "Bearer YOUR_TOKEN"
    }
  }'
The server will POST the complete analysis result to your webhook when ready.

​
Polling for Results

Check the analysis status using the job ID: 
curl http://localhost:8000/results/550e8400-e29b-41d4-a716-446655440000

​
Sync Mode

Sync mode blocks until analysis is complete and returns the full result immediately. Use this for simple scripts or when you need immediate results.

​
How It Works

  1. Submit analysis request to /analyze?sync=true
  2. Server blocks until AI analysis completes
  3. Receive complete analysis result (HTTP 200)

​
Example: Sync Analysis

curl -X POST "http://localhost:8000/analyze?sync=true" \
  -H "Content-Type: application/json" \
  -d '{
    "job_name": "test-pipeline",
    "build_number": 123,
    "ai_provider": "claude",
    "ai_model": "claude-sonnet-4-20250514"
  }'
Response: 
{
  "job_id": "550e8400-e29b-41d4-a716-446655440000",
  "job_name": "test-pipeline",
  "build_number": 123,
  "jenkins_url": "https://jenkins.example.com/job/test-pipeline/123/",
  "status": "completed",
  "summary": "3 failure(s) analyzed (2 unique error types)",
  "ai_provider": "claude",
  "ai_model": "claude-sonnet-4-20250514",
  "failures": [...],
  "base_url": "http://localhost:8000",
  "result_url": "http://localhost:8000/results/550e8400-e29b-41d4-a716-446655440000",
  "html_report_url": "http://localhost:8000/results/550e8400-e29b-41d4-a716-446655440000.html"
}

​
Choosing the Right Mode

Use Async When

  • Integrating with CI/CD pipelines
  • Analysis takes more than a few seconds
  • You need to show progress to users
  • Multiple analyses run concurrently

Use Sync When

  • Running quick scripts or one-off analyses
  • You need results immediately
  • Simplicity is more important than scalability
  • Testing or debugging the API

​
Implementation Details

Both modes use the same analysis engine under the hood. The difference is only in how results are delivered:
  • Async mode (main.py:366-451): Uses FastAPI’s BackgroundTasks to queue analysis jobs
  • Sync mode (main.py:384-423): Awaits the analysis coroutine directly before returning
main.py:372-376
if sync:
    logger.info(
        f"Sync analysis request received for {body.job_name} #{body.build_number}"
    )

    merged = _merge_settings(body, settings)
    ai_provider, ai_model = _resolve_ai_config(body)

    result = await analyze_job(
        body, merged, ai_provider=ai_provider, ai_model=ai_model
    )
Both modes support the same request parameters and produce identical analysis results.

Build docs developers (and LLMs) love

Get started for freeTalk to us