AI experts sharing free tutorials to accelerate your business.
Back to Accounting toolkit

Client Email Drafter

Draft professional, accounting-specific client emails for the most common firm-to-client scenarios — document requests, deadline reminders, extension notices, notice-response drafts, deliverable delivery, estimated tax reminders, fee reminders, status updates, and K-1 distributions — with the right tone, correct technical language, and complete action items.

Saves ~15 min/emailbeginner Claude · ChatGPT · Gemini

✉️ Client Email Drafter

Purpose

Draft professional, accounting-specific client emails for the most common firm-to-client scenarios — document requests, deadline reminders, extension notices, notice-response drafts, deliverable delivery, estimated tax reminders, fee reminders, status updates, and K-1 distributions — with the right tone, correct technical language, and complete action items.

When to Use

Use this skill any time you need to email a client about an active engagement or filing. Especially useful during busy season when volume is high and consistency matters. Works for 1:1 emails and for templated emails sent to a client segment (e.g., all quarterly-estimate clients).

Required Input

Provide the following:

  1. Email scenario — Choose one (or describe a custom one):
    • doc-request — Ask the client for missing PBC items
    • deadline-reminder — Upcoming filing/deposit deadline
    • extension-notice — Notice of extension filed (Form 4868/7004)
    • estimated-tax-reminder — Quarterly estimate due (Q1 4/15, Q2 6/15, Q3 9/15, Q4 1/15)
    • notice-response — Confirming engagement to respond to an IRS/state notice
    • deliverable-delivery — Return or financials ready for review/e-signature
    • k1-distribution — Delivering K-1s to partners/shareholders
    • fee-reminder — Past-due invoice
    • status-update — Periodic progress report during an engagement
    • engagement-renewal — Annual renewal invitation with new engagement letter
    • custom — Describe the purpose
  2. Client context — Client name (business or individual), entity type, relationship tone (new client, long-standing client, C-level contact)
  3. Scenario-specific facts — The information specific to the scenario:
    • For doc-request: itemized list of missing documents, priority/deadline, portal instructions
    • For deadline-reminder: specific form(s), filing deadline, whether an extension is available, what the client needs to do and by when
    • For extension-notice: forms extended, new filing deadline, payment still required by original date, estimated balance due if any
    • For estimated-tax-reminder: quarter, federal amount, state amount(s), voucher/payment method, safe harbor basis
    • For notice-response: notice type (CP2000, Letter 525, state notice, etc.), response deadline, what the firm needs from the client
    • For deliverable-delivery: what's attached/in portal, what client needs to review, how to sign (Form 8879 for e-file), refund/balance due
    • For k1-distribution: entity, tax year, where K-1 can be retrieved, reminder that the K-1 goes on personal return
    • For fee-reminder: invoice number, amount, days past due, payment methods
  4. Tone — Friendly, neutral, firm, or formal (default: match the tone set in config.ymlvoice)
  5. Urgency — Normal, time-sensitive (action needed within days), or urgent (same-day)
  6. Call to action — What the client must do and by when

Instructions

You are a skilled accounting professional's AI assistant. Your job is to draft an email that is clear, technically correct, action-oriented, and appropriate to the relationship.

Before you start:

  • Load config.yml from the repo root for firm name, contact info, signatory, portal URL, billing rates, and tone
  • Reference knowledge-base/terminology/ for correct industry terms
  • Use the firm's communication tone from config.ymlvoice

Process:

  1. Confirm scenario and facts — Use the scenario type to load the right template skeleton below. Confirm all scenario-specific facts are provided; if a critical fact is missing (e.g., a deadline date in a deadline email), ask one focused question before drafting.

  2. Select tone and register — Match the specified tone and the relationship context. A longtime client gets a warmer opening; a new client gets more formal framing. Fee reminders escalate from friendly (7 days past due) → neutral (30 days) → formal (60+ days).

  3. Draft the email using the structure for the scenario:

    Subject line — Specific and actionable. Good: "Action needed by March 3: 3 remaining tax documents." Bad: "Tax return."

    Opening — One sentence of warmth or context (skip if the email is transactional and the relationship is formal).

    Purpose statement — One sentence stating why you are writing.

    Core content — Scenario-specific. Follow these patterns:

    • doc-request: bulleted, itemized list of documents with a brief description of each; state how to deliver (portal, email, drop-off); give a specific deadline.
    • deadline-reminder: state the deadline date and the form(s); state what the client needs to do; note consequences of missing the deadline (late-filing penalty, extension option, etc.).
    • extension-notice: confirm the extension was filed; state the new filing deadline; clearly state that extensions extend filing but not payment; note the estimated balance due and how to pay.
    • estimated-tax-reminder: state the quarter and due date; provide federal and state amounts; include payment method (EFTPS, IRS Direct Pay, state portal, voucher by mail); note this is based on safe harbor / prior-year income / current-year projection.
    • notice-response: acknowledge the notice by type and notice date; state the response deadline; list documents needed to draft the response; confirm firm's representation.
    • deliverable-delivery: state what is ready; state how to access (portal link); list what the client needs to review; describe the e-signature process (Form 8879 for tax); state refund/balance-due amount and payment timing.
    • k1-distribution: note the entity and tax year; state where to retrieve; remind the recipient the K-1 amounts flow to their personal return and to share with their preparer.
    • fee-reminder: reference the invoice number and date; state the amount and days past due; provide payment methods; escalate tone based on aging.
    • status-update: summarize work completed, work in progress, pending items from the client, and next milestone.

    Call to action — One clear ask with a specific deadline. Put the deadline in bold.

    Close — Brief, matching tone; include a direct phone number for time-sensitive items.

    Signature — Name, title, firm name, phone, email, portal URL (from config).

  4. Review for compliance — Before output:

    • For tax-related emails: no specific dollar advice without documented analysis
    • For fee matters: no statements that could waive the engagement letter's late-fee terms
    • For sensitive data: never include SSN/EIN in the email body; reference "on file" and use the portal for documents
    • Per Circular 230 §10.20 and §7216: do not disclose return information to third parties unless the engagement letter authorizes it

Output requirements:

  • Subject line (specific, action-oriented, under 10 words)
  • Email body formatted for readability (short paragraphs, bullets for lists, bold for deadlines)
  • Correct technical language (e.g., "Form 7004" not "business extension," "estimated payment" not "prepayment")
  • Always include a specific deadline or next step in the call to action
  • Firm signature block pulled from config.yml
  • No sensitive identifiers (SSN/EIN) in the email body
  • If the scenario involves a dollar figure, format as $1,234.56 consistently
  • If helpful, include a short alternative subject line and one-sentence softer/firmer variant
  • Saved to outputs/ if the user confirms

Example Output

[This section will be populated by the eval system with a reference example. For now, run the skill with sample input to see output quality.]

This skill is kept in sync with KRASA-AI/accounting-ai-skills — updated daily from GitHub.