Hire an AI payroll hours and approval review

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• An exception list: each odd item, the row it came from and a question for HR
• A reconciliation of hours per person to the pay period
• A list of payroll questions with a reviewer and a due date
• An action register file you can open in a spreadsheet
• A hand-back summary with one next step

How it does the job well

• Payroll data is personal. Use employee IDs, not names, in every file you write. Leave out bank details, addresses and tax numbers.
• Raise every anomaly. Never correct, round or drop a number to make a total match.
• Do not connect to a payroll system, stage a payroll run or post a journal. Work from the files you were given.
• Do not invent a pay rule, an employee status or a tax treatment.
• Agree the question and the decision it supports before analysing.
• Check the data first: date range, grain, missing values, duplicates and definitions.
• Compare like with like: same period length, definitions and segments.

Mistakes it avoids

• Reporting a change without checking whether tracking or definitions changed.
• Treating correlation as cause.
• Averages that hide important segments or outliers.

What a good result looks like

An example, not your task: Payroll closes tomorrow and the timesheets are still coming in.

• Exception list. Each flagged item, such as a missing approval or duplicate time, with its source row and a question.
• Reconciliation. Hours per person in your source and in the working file, with any difference.
• Payroll questions. Each open question with a reviewer and a due date before the cut-off.
• Action register. A file with one row per task, with owner and date left blank if not named.
• Hand-back summary. Files saved, MISSING lines, checks missed and one next step.

How the work is checked

• Hours per person reconcile to the source period, or a difference is shown.
• No pay rule, employee status or tax treatment was made up.
• Every flag has a source row and a question.
• No name, bank detail or address appears in any file you wrote.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

Capabilities

• Payroll timesheet checker. Reads your export and flags missing approvals, duplicated time, missed punches and rate changes. Uses only the pay rules HR gave you. Hands the exception working file to the next step.
• Spreadsheet checker. Re-adds the period totals, checks employee IDs and formulas in the working file. Hands back the checked file and a list of any errors it found.
• Executive action register keeper. Turns each open question into a row with a reviewer and a due date before the cut-off. Leaves a name blank if you did not give one.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id check-payroll-inputs-before-the-cut-off only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/check-payroll-inputs-before-the-cut-off/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• The timesheet export for the pay period
• Approved leave for the same period
• The confirmed roster and pay rates
• Pay period start and end dates, and the pay date
• Pay rules from HR, such as overtime and rounding
• Last period's totals (optional, but needed to spot big jumps)
• The cut-off date and the names of your reviewers (optional)

Helpful context (optional)

• Last period's exception list.
• A named reviewer at HR or your payroll provider.

Tools and access

• A spreadsheet tool such as Excel or Google Sheets: Lets you open the working file and re-check totals. The AI cannot recalculate a workbook without one.

Functions and deliverables

• Payroll hours and approval review: Check payroll source hours, approvals and totals. Get a list of issues and questions for the right reviewers before the cutoff.
  • An exception list: each odd item, the row it came from and a question for HR
  • A reconciliation of hours per person to the pay period
  • A list of payroll questions with a reviewer and a due date
  • An action register file you can open in a spreadsheet
  • A hand-back summary with one next step
• Payroll timesheet checker: Bring timesheets and pay dates. Get hours by person and a run sheet with flagged issues.
  • Hours totalled by person
  • List of flagged problems
  • Run staged for line-by-line approval
  • Journal entry for the books
  Source: Anthropic.
• Spreadsheet checker: Check a spreadsheet for odd totals, broken formulas and messy data. Your original stays safe, and you get findings with proof and a new file.
  • A sheet inventory and scope
  • Findings with evidence locations
  • Data quality and formula issues
  • A new output file with your edits
  • Reconciliation results
  Source: Burhan Sebin.
• Action item tracker: Share meeting notes or email threads. Get action items with owners, deadlines and unresolved commitments.
  • action-register.csv: id, task, owner, due_at, timezone, status, depends_on, source_id
  • clarifications.md
  • proposed-task-changes.md
• Metric definition writer: Share a metric, calculation and business use. Get a YAML definition, validation results and questions for the data owner.
  • A filled metric, dimension or entity YAML
  • A validation report
  • dbt-ready field names
  • Notes on edge cases
  Source: Nimrod Fisher.

Tasks you can use it for

• Payroll closes tomorrow and the timesheets are still coming in.
• A manager says some hours were never approved.
• Total pay looks far higher than last period and you do not know why.
• Someone started or left partway through the pay period.

Limits to check before starting

• You want payroll run or pay changed. This only flags. HR or your payroll provider makes changes.
• You need tax or employee status advice. Ask your payroll provider or accountant.
• You want one quick look at a single timesheet. Use the Payroll timesheet checker.
• You are sorting business expenses. Use the Bookkeeping expense matching checker workflow.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Download the handoff pack. It is a zip with HANDOFF.md and the skill files this workflow uses.
• Give the folder to Claude, ChatGPT or Gemini and say "Follow HANDOFF.md".
• Attach the timesheet export, leave, roster and HR rules. Use employee IDs instead of names where you can.
• Read the exception list first. Check each flagged row against your own records.
• Send the open questions to HR or your payroll provider before the cut-off. They decide any change.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/check-payroll-inputs-before-the-cut-off for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.