DeepRead Pay Stubs#

Turn pay stubs and earnings statements into clean, typed JSON — employer, employee, pay period, gross and net pay, tax withholdings, deductions, and year-to-date totals — with a needs_review flag on every field. Built for income verification, where a misread number has real consequences.

This skill instructs the agent to POST documents to https://api.deepread.tech and poll for results. No system files are modified.

What You Get Back#

json

{
  "schema_version": "dp02",
  "status": "completed",
  "extraction": {
    "fields": [
      {"key": "employer_name", "value": "Acme Corp", "needs_review": false, "location": {"page": 1}},
      {"key": "employee_name", "value": "Jordan Rivera", "needs_review": false, "location": {"page": 1}},
      {"key": "pay_period_start", "value": "2026-03-16", "needs_review": false, "location": {"page": 1}},
      {"key": "pay_period_end", "value": "2026-03-31", "needs_review": false, "location": {"page": 1}},
      {"key": "pay_date", "value": "2026-04-04", "needs_review": false, "location": {"page": 1}},
      {"key": "gross_pay", "value": 3520.00, "needs_review": false, "location": {"page": 1}},
      {"key": "net_pay", "value": 2614.18, "needs_review": false, "location": {"page": 1}},
      {"key": "ytd_gross", "value": 21120.00, "needs_review": false, "location": {"page": 1}},
      {"key": "deductions", "value": [
        {"type": "Federal Tax", "amount": 528.00},
        {"type": "Social Security", "amount": 218.24},
        {"type": "Health Insurance", "amount": 159.58}
      ], "needs_review": false, "location": {"page": 1}}
    ]
  }
}

Setup#

bash

open "https://www.deepread.tech/dashboard/?utm_source=clawhub"
export DEEPREAD_API_KEY="sk_live_your_key_here"

Schema#

json

{
  "type": "object",
  "properties": {
    "employer_name":   {"type": "string", "description": "Employer name"},
    "employee_name":   {"type": "string", "description": "Employee name"},
    "pay_period_start":{"type": "string", "description": "Pay period start (YYYY-MM-DD)"},
    "pay_period_end":  {"type": "string", "description": "Pay period end (YYYY-MM-DD)"},
    "pay_date":        {"type": ["string","null"], "description": "Pay/check date (YYYY-MM-DD)"},
    "pay_frequency":   {"type": ["string","null"], "description": "Weekly, biweekly, semimonthly, monthly"},
    "gross_pay":       {"type": "number", "description": "Gross pay this period"},
    "net_pay":         {"type": "number", "description": "Net (take-home) pay this period"},
    "ytd_gross":       {"type": ["number","null"], "description": "Year-to-date gross pay"},
    "ytd_net":         {"type": ["number","null"], "description": "Year-to-date net pay"},
    "deductions": {
      "type": "array",
      "description": "Itemized deductions and taxes",
      "items": {"type": "object", "properties": {
        "type":   {"type": "string", "description": "Deduction name (Federal Tax, 401k, Health, ...)"},
        "amount": {"type": "number", "description": "Amount this period"}
      }, "required": ["type", "amount"]}
    }
  }
}

Verify Income (Python)#

python

fields = {f["key"]: f["value"] for f in result["extraction"]["fields"]}
gross = fields["gross_pay"]
freq = (fields.get("pay_frequency") or "").lower()
periods = {"weekly":52,"biweekly":26,"semimonthly":24,"monthly":12}.get(freq)
if periods:
    print(f"Estimated annual gross: ${gross * periods:,.0f}")
# Sanity-check: net + sum(deductions) should ≈ gross
ded = sum(d["amount"] for d in fields.get("deductions", []))
if abs((fields["net_pay"] + ded) - gross) > 1.0:
    print("⚠ net + deductions != gross — review extraction")

Use Cases#

• Lending / mortgage — verify applicant income from recent pay stubs
• Rental / tenant screening — confirm income-to-rent ratios
• HR / payroll audits — reconcile stubs against payroll runs
• Gig / benefits eligibility — compute annualized income

Redact Before Sharing#

Pay stubs carry names, partial SSNs, and bank details. Redact with deepread-pii before sending externally: clawhub install uday390/deepread-pii

Tips#

• Compute the annualization in code (shown above) — don't ask the model to guess pay frequency math.
• Recurring employer format? Build a blueprint at https://www.deepread.tech/dashboard/optimizer.
• Check needs_review — flagged amounts are the only ones to eyeball.

Related DeepRead Skills#

• deepread-tax-forms — W-2/1099 income verification — clawhub install uday390/deepread-tax-forms
• deepread-bank-statements — income deposits from statements — clawhub install uday390/deepread-bank-statements
• deepread-pii — redact sensitive data — clawhub install uday390/deepread-pii

Support#

• Dashboard: https://www.deepread.tech/dashboard
• Email: support@deepread.tech

---

Get started free: https://www.deepread.tech/dashboard/?utm_source=clawhub