Install any skill in seconds. Free to start, no credit card required.
Get Started Free →SSH into a server, restart Nginx, and verify it's back up
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -48% | 0% |
| Name | Type | Required | Description | |------|------|----------|-------------| | host | string | Yes | Target server hostname or IP |
Before starting execution, initialize task_context.json:
json{ "task_id": "<task_id from input>", "current_step": 0, "current_step_id": null, "status": "running", "steps": { "restart": "pending", "verify": "pending" }, "updated_at": "<ISO timestamp>" }
Update this file after each step completes. On error, set step status to "failed" and overall status to "failed".
Type: agent Description: SSH in and restart Nginx
Launch an independent agent with the following prompt file:
Dispatch instruction:
SSH into {host} and run systemctl restart nginx
Agent workflow:
restart_result.jsonAfter completing this step, update task_context.json:
current_step_id to "restart"steps.restart to "completed"Type: agent Description: Verify Nginx is healthy
restart_result.json (from Step restart, schema: schemas/restart.schema.json)Launch an independent agent with the following prompt file:
Dispatch instruction:
curl http://{host}/health and check the status code
Agent workflow:
schemas/restart.schema.jsonverify_result.jsonAfter completing this step, update task_context.json:
current_step_id to "verify"steps.verify to "completed"Other measured skills in the registry, with their headline benchmark lift.