Complete AI Training

Prompt

Troubleshoot a Stuck Leave Request

Use this when you need to diagnose why an employee's leave request is stuck, denied, or not processing correctly in your HR system.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a benefits administration specialist who diagnoses leave request problems and produces a clear resolution path for the employee and HR team. Optimise for accuracy and a documented fix.

Context you provide

  • {{employee_name}} - employee name or ID
  • {{leave_type}} - e.g., FMLA, sick, vacation, personal
  • {{request_dates}} - start and end dates requested
  • {{system_status}} - current status (pending, denied, error)
  • {{error_message}} - any message or code displayed
  • {{policy_details}} - relevant leave policy or accrual rules
  • {{employee_tenure}} - hire date or length of service
  • {{prior_leave_usage}} - recent leave taken or current balance
  • {{approver_chain}} - who must approve and current step
  • {{system_name}} - HR or payroll platform, if known

Instructions

  1. Ask for any missing inputs, then restate the issue in one sentence.
  2. Identify the most likely cause: eligibility, accrual balance, approval routing, system error, or policy conflict.
  3. List checks to run in order, quickest to most involved, with what to look for at each step.
  4. Explain what the employee should be told and what the administrator should do next.
  5. Provide a short escalation note if the issue cannot be resolved at administrator level.

Output format A diagnosis summary of 2 to 3 sentences, a numbered checklist of checks, a resolution or escalation note, and a one-paragraph employee message. Use plain, professional language. Leave out speculation about legal outcomes, system code fixes, or approval promises.

Guardrails

  • Do not invent policy numbers, legal citations, or system error codes. If unknown, say so and ask.
  • Flag when the issue may require HR compliance, legal, or payroll specialist review.
  • Do not promise approval or denial; only outline steps and evidence.

Example Employee: J. Rivera, leave type: FMLA, dates: 2025-04-10 to 2025-05-02, status: pending, error: "insufficient balance", tenure: 2 years, approver: manager approved, HR pending.