Skip to main content
    Use-case library
    Tax

    Tax Provision Support

    Draft provision workpapers and reconcile book-to-tax differences.

    LLM workflowAutonomy: DraftQualityRisk & ControlMedium effortOrg maturity: AugmentedCFO
    What changed

    Rules pull the trial balance and compute the current and deferred provision. A language model proposes the book-to-tax classification of each difference for the tax team to confirm and explains movements versus the prior period, leaving the team to validate judgment items rather than build schedules.

    How it's built

    LLM workflow (rung 4 of 6). A coded control path: chain, router, parallel fan-out, evaluator-optimizer.

    Not an agent: the sequence of steps is fixed in advance, so the whole path can be tested end to end.

    A language model handles steps 2 and 4. Rules: steps 1 and 3 · Human: step 5.

    Autonomy: T1 · Draft
    Produces a complete output for a human to review and release. Human role: reviews every item and releases it.
    How it is governed
    The path is testable end to end. Each step is independently evaluable.
    What the certification exams call this
    • Anthropic (Claude Certified Architect – Foundations (CCAR-F))Workflow: models and tools on predefined code paths (chaining, routing, parallelization, evaluator-optimizer). source
    • OpenAI (OpenAI agent guidance (Agents SDK, Codex, OpenAI Academy))Orchestrating via code, which the Agents SDK calls more deterministic and predictable. source
    • Google Cloud (Generative AI Leader)Workflow agents (sequential, parallel, loop) that run steps in a fixed order. source
    • Microsoft (AI-901 (Azure AI Fundamentals) and AB-100 (Agentic AI Business Solutions Architect))Sequential and concurrent orchestration patterns, with steps in a set order. source

    How Koko classifies AI · the full ladder is taught in AI & Agentic Delivery Foundations.

    Try it

    Illustrative demo

    Worked example of "Tax Provision Support" on synthetic data — edit the inputs to tailor it. Edit the inputs to tailor the output — it's baked synthetic data, so nothing leaves your browser until you run it live.

    Synthetic demo company (a ~$5B high-growth US AI SaaS) — swap in any name to tailor the output.

    What this run should concentrate on.

    Run live in Claude

    Workflow

    1. 1.RulesPull the trial balance, prior provision and known adjustments.
    2. 2.Generative AIPropose the book-to-tax classification of each difference (permanent or temporary) for the tax team to confirm.
    3. 3.RulesCompute the current and deferred provision build.
    4. 4.Generative AIExplain movements vs. prior period.
    5. 5.HumanRoute workpapers to tax for review; escalate judgment items.

    Prompt recipe

    From the attached trial balance and prior provision, draft the book-to-tax difference schedule, classify each difference (permanent/temporary), build the current and deferred provision, and explain the period-over-period movement. Flag any item requiring a judgment call rather than deciding it. Run every calculation in code (a code-execution or data-analysis tool, or a spreadsheet), never in prose, and show the table it produced.
    Try in Ask KokoAI
    Controls required
    • Source data reconciled to the ledger
    • Tax reviews and approves; judgment items escalated, not auto-decided
    • Workpaper traceability retained for audit
    • Regulated/confidential data handling
    HomeNewsLearnIntelBuildAsk Koko