ai.teampredict/hr-calculators

None of its 22 tools are described as changing anything. For 22 the reading was uncertain, so no mark is shown.

• employee_turnover_cost

  “Estimate what employee turnover costs a company per year, and how much cutting turnover would save.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  5 inputs, 3 required

  • employeesintegerrequired“Number of employees (headcount), a whole number from 1 to 1,000,000.”
  • average_salary_usdnumberrequired“Average annual salary per employee, in US dollars.”
  • annual_turnover_rate_percentnumberrequired“Share of employees who leave in a year, as a percentage from 0 to 100.”
  • replacement_cost_percent_of_salarynumber“What replacing one person costs, as a percentage of their annual salary.”
  • target_reduction_pointsnumber“Percentage points of annual turnover you want to cut, for the savings figure.”
• employee_turnover_rate

  “Calculate the employee turnover (attrition) rate for a month, quarter or year from starting headcount, ending headcount and the number of people who left.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  4 inputs, 3 required

  • periodstring“The length of the measurement period: month, quarter or year.”
  • starting_headcountintegerrequired“Employees on the payroll at the start of the period.”
  • ending_headcountintegerrequired“Employees on the payroll at the end of the period.”
  • separationsintegerrequired“People who left during the period, voluntarily or not.”
• employee_retention_rate

  “Calculate the employee retention rate exactly, by separating leavers who were on the payroll at the start of the period from leavers who were hired during it.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  4 inputs, 4 required

  • starting_headcountintegerrequired“Employees on the payroll when the period opened.”
  • new_hiresintegerrequired“People hired during the period.”
  • leavers_from_starting_groupintegerrequired“Leavers who were already on the payroll when the period opened.”
  • leavers_from_new_hiresintegerrequired“Leavers who were hired during the period.”
• cost_of_a_bad_hire

  “Estimate the total cost of one bad hire: the salary paid for lost productivity plus recruiting, onboarding, manager time, severance and the cost of hiring a replacement.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  8 inputs, 2 required

  • annual_salary_usdnumberrequired“Annual salary of the role, in US dollars.”
  • months_employednumberrequired“Months the hire stayed before leaving or being let go, 0 to 120.”
  • lost_productivity_percentnumber“Share of the salary paid during the tenure that bought no useful work, 0 to 100.”
  • recruiting_cost_usdnumber“Recruiting and agency cost, in US dollars.”
  • onboarding_cost_usdnumber“Onboarding and training cost, in US dollars.”
  • manager_time_cost_usdnumber“Value of manager and team time spent on the hire, in US dollars.”
  • severance_and_exit_cost_usdnumber“Severance and exit cost, in US dollars.”
  • rehire_cost_usdnumber“Cost of finding and hiring the replacement, in US dollars.”
• enps_score

  “Calculate an employee Net Promoter Score (eNPS) from survey response counts: the percentage of promoters minus the percentage of detractors, from -100 to +100.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  3 inputs, 3 required

  • promotersintegerrequired“Number of responses scoring 9 or 10.”
  • passivesintegerrequired“Number of responses scoring 7 or 8.”
  • detractorsintegerrequired“Number of responses scoring 0 to 6.”
• fte_calculator

  “Count full-time equivalent employees (FTE) from full-time staff and groups of part-time staff, and run the Affordable Care Act count that decides whether an employer has 50 or more full-time equivale…”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  3 inputs, 1 required

  • full_time_employeesintegerrequired“Number of full-time employees.”
  • full_time_hours_per_weeknumber“Weekly hours of one full-time schedule at your company.”
  • part_time_groupsarray“Up to 6 groups of part-time staff, each with a head count and weekly hours per person.”
• pto_accrualon 1 other server

  “Work out how paid time off accrues: hours earned per pay period, per month and per hour worked, plus the balance after each pay period, with an optional balance cap and hours already used.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  9 inputs, 1 required

  • annual_ptonumberrequired“Paid time off granted per year, in the unit given by unit (for example 15 days or 120 hours).”
  • unitstring“Whether annual_pto is in days or hours.”
  • hours_per_daynumber“Hours in one working day, used to convert days to hours.”
  • pay_frequencystring“How often accrual is credited: weekly (52 a year), biweekly (26), semimonthly (24) or monthly (12).”
  • hours_worked_per_yearnumber“Paid hours worked in a year, for the accrual per hour worked.”
  • starting_balance_hoursnumber“PTO balance in hours at the start of the year.”
  • hours_used_this_yearnumber“PTO hours already taken this year.”
  • pay_periods_completedinteger“Pay periods already paid this year, so the balance can be shown as of today.”
  • balance_cap_hoursnumber“Maximum PTO balance in hours; accrual pauses at the cap.”
• salary_increase

  “Calculate a pay raise: the new annual and hourly pay, the raise per year, month and biweekly paycheck, the raise as a percentage, and whether it keeps up with inflation.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  6 inputs, 1 required

  • pay_unitstring“Whether current_pay and new_pay are per year or per hour.”
  • current_paynumberrequired“Current pay in US dollars, per year or per hour according to pay_unit.”
  • raise_percentnumber“The raise as a percentage of current pay, for example 4 for 4%.”
  • new_paynumber“The new pay in US dollars, per year or per hour according to pay_unit.”
  • hours_per_weeknumber“Hours worked per week, to move between hourly and yearly pay (52 weeks a year).”
  • inflation_percentnumber“Expected yearly inflation as a percentage, to show the raise in real terms.”
• severance_pay

  “Estimate a severance package: weeks of pay per year of service times years of service times weekly pay, with an optional plan minimum and maximum, plus unused PTO paid out, employer-paid health cover…”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  9 inputs, 3 required

  • annual_salary_usdnumberrequired“Annual salary in US dollars; weekly pay is salary divided by 52.”
  • years_of_servicenumberrequired“Years of service, decimals allowed (for example 4.5).”
  • weeks_of_pay_per_yearnumberrequired“Weeks of pay the plan gives for each year of service.”
  • minimum_weeksnumber“The plan's minimum severance in weeks of pay, if any.”
  • maximum_weeksnumber“The plan's cap in weeks of pay, if any.”
  • unused_pto_daysnumber“Unused PTO days paid out, valued at salary divided by 260 working days.”
  • health_coverage_monthsnumber“Months of health coverage the employer pays.”
  • monthly_health_premium_usdnumber“Monthly cost of that coverage in US dollars.”
  • other_lump_sums_usdnumber“Any other lump sums, such as a pro-rated bonus, in US dollars.”
• us_paycheck_take_home

  “Calculate US take-home (net) pay from a salary or an hourly wage for 2026, after federal income tax (from the Form W-4), Social Security, Medicare and the state's own income tax and payroll contribut…”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  19 inputs, 1 required

  • statestringrequired“Two-letter postal code of the state where the employee works and lives: all 50 states and DC (for example CA, NY, TX, DC).”
  • pay_typestring“Whether the employee is paid a yearly salary or an hourly wage.”
  • annual_salary_usdnumber“Yearly gross salary in US dollars.”
  • hourly_rate_usdnumber“Hourly wage in US dollars.”
  • regular_hours_per_weeknumber“Regular (straight-time) hours per week, hourly pay only.”
  • overtime_hours_per_weeknumber“Overtime hours per week paid at time and a half, hourly pay only.”
  • pay_frequencystring“How often the employee is paid: weekly (52 paychecks a year), biweekly (26), semimonthly (24, twice a month) or monthly (12).”
  • filing_statusstring“Filing status on the federal Form W-4, step 1(c): single (or married filing separately), married (filing jointly) or head (head of household).”
  • w4_step2_checkboxboolean“The 'multiple jobs or spouse works' box in Form W-4 step 2(c).”
  • w4_children_under_17integer“Qualifying children under 17 for Form W-4 step 3 (a $2,200 credit each in 2026).”
  • w4_other_dependentsinteger“Other dependents for Form W-4 step 3 (a $500 credit each).”
  • w4_other_income_usd_per_yearnumber“Other income not from jobs, Form W-4 step 4(a), in US dollars per year.”
  • w4_deductions_usd_per_yearnumber“Deductions above the standard deduction, Form W-4 step 4(b), in US dollars per year.”
  • w4_extra_withholding_usd_per_paychecknumber“Extra federal withholding per paycheck, Form W-4 step 4(c), in US dollars.”
  • traditional_401k_percentnumber“Pre-tax 401(k) contribution as a percentage of gross pay, 0 to 100 (capped at the 2026 IRS limit).”
  • health_premiums_usd_per_paychecknumber“Pre-tax health, dental and vision premiums per paycheck, in US dollars.”
  • hsa_usd_per_paychecknumber“Pre-tax HSA contribution through payroll per paycheck, in US dollars (capped at the 2026 family limit).”
  • state_allowancesinteger“Allowances or exemptions on the state's own withholding form, for states that have one.”
  • local_taxstring“Id of a city or county income tax to include (for example 'nyc' or 'yonkers' for New York).”
• us_bonus_tax

  “Calculate how much tax is withheld from a US bonus in 2026 and what is left: federal income tax (the 22% flat rate or the combined method, 37% above $1 million), Social Security, Medicare, and the st…”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  9 inputs, 3 required

  • statestringrequired“Two-letter postal code of the state where the employee works and lives: all 50 states and DC (for example CA, NY, TX, DC).”
  • annual_salary_usdnumberrequired“Regular yearly salary in US dollars, before the bonus.”
  • bonus_usdnumberrequired“Gross bonus in US dollars.”
  • pay_frequencystring“How often the employee is paid: weekly (52 paychecks a year), biweekly (26), semimonthly (24, twice a month) or monthly (12).”
  • filing_statusstring“Filing status on the federal Form W-4, step 1(c): single (or married filing separately), married (filing jointly) or head (head of household).”
  • w4_step2_checkboxboolean“The 'multiple jobs or spouse works' box in Form W-4 step 2(c).”
  • bonus_paidstring“Whether the employer pays the bonus as a separate payment (the usual case) or adds it to a regular paycheck.”
  • federal_methodstring“How the employer withholds federal income tax on a bonus paid separately: the flat 22% rate (the common choice) or the combined (aggregate) method.”
  • earlier_supplemental_pay_usdnumber“Bonuses and other supplemental pay already paid this calendar year, before this one, in US dollars.”
• hourly_to_salary

  “Convert pay between hourly, daily, weekly, biweekly, semimonthly, monthly and yearly, in either direction, from one amount.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  5 inputs, 1 required

  • amount_usdnumberrequired“The pay you know, in US dollars for the period named by period.”
  • periodstring“The period that amount_usd is paid per: hour, day, week, biweekly (every two weeks, 26 a year), semimonthly (twice a month, 24 a year), month or year.”
  • hours_per_weeknumber“Hours worked in a paid week.”
  • days_per_weeknumber“Days worked in a week, for the per-day figure.”
  • paid_weeks_per_yearnumber“Paid weeks in a year; use 50 or 48 to leave out unpaid vacation.”

Each quoted sentence is the first sentence of the publisher’s own description. The kind of tool and the “changes data” mark are our AI model’s reading of that description, not something we tested: two differently worded runs had to agree, otherwise no label is shown, and nobody has checked them by hand. A tool without the “changes data” mark is described as only reading, searching or calculating, or the two runs did not agree. No tool was called. Read the publisher’s full description before relying on it. “Publisher declares” lists the hints the server itself sends with the tool (MCP annotations: read-only, destructive, idempotent, open-world); they are the publisher's claims, not checked by us. Inputs are the parameters in the tool's own input schema: name, declared type, whether it is required, and the first sentence of its description. Read 2026-10-26.

The registry verifies a package once, when it is published, and does not re-check it. A difference below is about the listing, not a judgement of the publisher.