Quickstart
Get started with Knify in under 5 minutes. This guide will walk you through creating your first agentic job.
Prerequisites
Before you begin, you'll need:
- A Knify API key (contact your administrator or see the API keys section)
curlor any HTTP client- (Optional) Your agent's workspace files
Your First Job
Set Your API Key
All API requests require authentication via the Authorization header:
export KNIFY_API_KEY="your_api_key_here"
Create a Simple Job
Let's start with a basic Cursor task that writes a Python script:
curl -X POST https://api.knify.io/jobs \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $KNIFY_API_KEY" \
-d '{
"spec": {
"job_type": "cursor_task",
"prompt": "Write a Python script that calculates fibonacci numbers"
}
}'
Response:
{
"job_id": "job_abc123",
"status": "running"
}
Check Job Status
Monitor your job's progress:
curl https://api.knify.io/jobs/job_abc123 \
-H "Authorization: Bearer $KNIFY_API_KEY"
Response includes:
- Current status (
pending,running,completed,failed) - Execution events and logs
- Sandbox ID (for continuation)
- Error details (if failed)
Get Job Results
Once the job completes, fetch the artifacts:
curl https://api.knify.io/jobs/job_abc123/artifacts \
-H "Authorization: Bearer $KNIFY_API_KEY"
Artifacts include:
- Generated files
- Execution logs
- Agent output
- Any reports or data produced
Continue the Conversation
Jobs can be continued to build multi-turn workflows:
curl -X POST https://api.knify.io/jobs/job_abc123/continue \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $KNIFY_API_KEY" \
-d '{
"spec": {
"job_type": "cursor_task",
"prompt": "Now run the fibonacci script and show me the output"
}
}'
The agent resumes in the same sandbox with all files and context preserved.
Next Steps
API Endpoint
- Base URL:
https://api.knify.io(or your self-hosted instance) - Authentication:
Authorization: Bearer <api_key> - Documentation: See the API Reference section for full endpoint details