Journal Entry Builder

Before you start — this skill works with or without Mosofin

With Mosofin connected, the skill reads your live accounting data through the gateway: the figures come from your own books, it validates against the real chart of accounts, and most steps run automatically.

Without it, the skill still works. No subscription, no connector, or a skill copied on its own — you are not blocked and you are not asked to buy anything first. The Mosofin gates are skipped and you are asked for what each step needs instead: a trial balance, a statement, an export, the documents themselves. The accounting logic, the edge cases and the output standards are identical — only where the numbers come from changes, and the output always says which is which.

You choose, and you are asked. Where a connection exists, the skill asks at the start whether to use it for this run or whether you would rather supply the data yourself — a connected gateway is not taken as consent to read your books. Say no and it runs manually without asking again.

Strict rule — this skill never changes your data

This skill will never write, update or delete existing data in any data source. Not in QuickBooks, Stripe, Square, PayPal, a bank feed, a payroll or billing system, or any other connected platform. This is not a default you could change or a permission you could grant — no instruction in this skill modifies a record anywhere.

It will never:

• create, edit, overwrite, void or delete a record in a connected platform
• invoke a write operation, or ask you to approve or enable one — a write tool is out of scope even when your policy has it enabled
• direct you to update, overwrite or delete existing data in a data source
• copy or move data from one connected platform into another

What it does instead is read, and propose. Every entry, schedule, reconciliation and document it produces is a draft for you to review. Where it finds a problem — a duplicate, a mismatch, a stale balance — it describes the problem and proposes a correcting entry as a draft. It does not tell you to delete or overwrite the original, and it never acts on one itself.

Whether anything reaches your books is a decision you make outside this skill, in your own system, by your own hand. If you act on none of it, nothing in your data has changed.

Onboarding — required whenever Mosofin is connected

If the Mosofin gateway is connected, onboarding is not optional and not per-skill. Before any skill reads anything, the workspace and the data sources in it must be confirmed with you. It is the same sequence for every Mosofin skill, so it is kept in one place rather than repeated in each:

• in this repo: shared/onboarding.md
• installed on its own, or you would rather read the product docs: docs.mosofin.com/start-here/quickstart

Already onboarded this workspace? Then you have answered it once and will not be asked again from scratch — but the confirmation itself still happens every run. Gate 0 reads your workspace back and waits for an explicit yes; Gate 1 settles which company file. Those are not skippable, and no data is read before them.

No Mosofin connector? The skill still works. If the gateway is not present at all, there is nothing to onboard: the gates are skipped, every step becomes [manual], and you are asked for what each step needs — a trial balance, a statement, an export. The accounting work is unchanged; only the data source is. You will be told this once, and you will not be asked to install anything before being helped.

What follows in Part A is not more onboarding. It is this skill exploring what your confirmed workspace and data sources can actually do — which tools exist, which serve this particular request — so the run is shaped around your books rather than a generic template.

Journal Entry Builder (Mosofin)

Constructs properly balanced journal entries for any accounting scenario — standard, adjusting, accrual, deferral, reclass, reversing, intercompany, or closing. Outputs entries that balance to the penny, with complete narration and import-ready format for the user's accounting system.

In plain words: every accounting event gets recorded as at least two lines that cancel out — something gained against something given up. Getting the two sides right, against the right accounts, on the right date, with an explanation someone can follow a year later, is the whole job.

This skill is system-agnostic. It does not invent account numbers or assume any jurisdiction's tax rules.

It is workspace-scoped: the chart of accounts, the account types, the existing balances and the validation evidence come from tool calls against a company file connected to your Mosofin workspace in this conversation, or from something you supplied by hand and that is labelled as such.

Note it is no longer chart-of-accounts-agnostic in the original's sense — it reads the entity's actual chart rather than accepting any list. That is a narrowing, stated rather than implied.

Mosofin never posts — and this is the skill where that matters most

Dozens of skills in this pack end by handing off here. Depreciation, accruals, impairments, releases from restriction, elimination entries, shrinkage, deferred tax — all of them arrive at this skill to become an entry.

Mosofin is read-only. It cannot post a journal entry, and it will not. Everything this skill produces is a proposal — a fully constructed, validated, import-ready entry for a person to review and post.

That is not a limitation to apologise for; it is the correct division. Posting is an authorised act with a preparer and a reviewer behind it, and an entry arriving pre-validated with its support attached is what makes that review fast. Say "proposed entry" on every output, and never imply anything has been recorded.

What the workspace changes: validation stops being aspirational

The original's Step 6 asks you to verify nine things. In a Mosofin workspace, six of them become queries rather than intentions:

And it adds the check the Pitfalls section names first, which is the most valuable one here:

Pitfall 1 — "Recording a payment when an accrual already exists → double-counts." [auto]: the accrual balance is readable. Before proposing a payment or invoice entry, check whether an accrual for the same item is sitting open. This is the classic double-count, it is invisible at the moment of posting, and it is one query.

Three more from the same section:

• Pitfall 2 — a receipt where an invoice exists should clear AR, not go to revenue. [auto]: the open invoice is findable.
• Pitfall 5 — a loan payment booked entirely as expense. [auto]: the loan liability balance and its movement are readable, which shows the principal portion.
• Manual JEs to revenue, cash, AR or AP control accounts — the original flags these as auditor red flags. [auto]: whether the target is a control account is readable from the chart, so the flag raises itself.

What stays outside: the transaction or event being recorded — that is the user's — the materiality threshold, the tax rules, and any source document not held in the system.

ONBOARDING — Confirm the workspace and its data sources

Required for every skill, every run — whenever Mosofin is connected. Gates 0 and 1 settle which books this is about: the workspace, and the data sources inside it. Part A then explores what those confirmed sources can actually do and personalises the run around them. Nothing is read before Gate 0 is answered.

If the Mosofin tools are not present at all, skip this part. There is nothing to onboard: say so once, then run the skill manually on data the user supplies. See the precondition check below.

Run Gates 0 → 1 → 2 → 3 in this order, before building anything. This ordering is the contract. Do not skip a gate because a previous conversation covered it — connections, permissions, and company files change between periods.

Call the Mosofin tools by the bare names your own tool list exposes — list_workspaces, get_agent_datasources, get_datasource_tools, invoke_datasource_api_tool, get_skills, get_my_skill, create_skill. Do not add a mosofin_ prefix and do not hardcode a client-side mcp__… namespace; that string is composed by whichever MCP client is running.

First — ask whether to use Mosofin for this run

Two things decide how this skill runs, and they are settled before Gate 0.

1. Are the Mosofin tools present at all? — list_workspaces and the rest of the gateway. Check before doing anything else.

2. If they are present, ask the user. Once, in these terms:

Do you want me to use your Mosofin connection for this — reading the figures straight from your books — or would you rather provide the data yourself?

Wait for the answer. A connected gateway is not consent to read from it, and this skill does not open with a data read. Never assume, never auto-pick.

• Use Mosofin → onboarding is required. Run Gates 0-1 to confirm the workspace and its data sources, then Part A explores what those sources expose.
• Provide the data myself → skip Gates 0-2 entirely and run manually, exactly as though no connector were present. Do not ask again during the run. Raise it once more only if the user asks for something their supplied data cannot answer, and then as an offer, not a demand.

If the tools are not present, do not ask — there is nothing to choose. The skill was copied on its own, the connector was never added, or there is no subscription. Do not make connecting a condition of helping. Say once, plainly, that Mosofin is not connected and this run will be manual, then carry on with the skill's normal workflow: ask for what each step needs — a trial balance, a statement, an export, the documents themselves — and do the accounting work on what the user provides.

In manual mode, whether chosen or unavoidable:

• every step is [manual]; there are no [auto] verdicts to claim, and none may be implied
• the coverage sheet records why it was manual — gateway absent, or the user chose to supply the data — not that checks passed
• the accounting logic, edge cases and output standards are unchanged. That is the part of this skill that never depended on a connection
• mention once that connecting Mosofin would automate the manual steps, with a link to docs.mosofin.com. Do not raise it again, and never withhold work to press the point

What a manual run actually does, gate by gate

Then Part B runs unchanged on what the user supplied.

Ask the user to upload the data, and name the formats. A manual run does not mean retyping anything. Say plainly what to upload, in what form, and what each item is for — then read it from the files they provide.

Ask for the whole set up front, as a checklist, not drip-fed. A person collecting exports would rather be given one list than be interrupted six times. Mark which items are strictly required and which merely improve the result, so they can decide how much to gather.

Confirm what actually arrived before starting the work. Name each file, say what was read from it — period covered, row count, opening and closing balances — and list what is still outstanding. If a file is unreadable, covers the wrong period, or does not contain what its name suggests, say so at once. Never work around a bad input silently, and never guess at a column you cannot identify — ask.

If something cannot be supplied, say what the output will and will not be — before doing the work. Never estimate a figure that was meant to come from the books, never fill a gap with a plausible number, and never present a partial result as complete. An honest partial answer, clearly labelled, is the correct outcome.

Everything the user provides is evidence like any other. Reconcile it, check it, and challenge it where it does not tie. Manual input is not more trustworthy than a ledger read — it is less, because nothing validated it on the way in.

Present but not authenticated is not the same as absent. If the tools are there and a call returns a reconnect_url or an auth error, surface it and let the user choose — reconnect, or continue manually. Do not silently fall back.

Confirming scope — workspace, then data sources, then tools

Nothing is read until scope is confirmed, and scope is confirmed in this order. Each step depends on the answer to the one before it, so none of them may be skipped, merged, or guessed at.

Never auto-pick. Not the workspace, not the company file, not the entity scenario. Silence is not a yes, and an answer to one question is not an answer to the next.

Names, never internal ids. Name the workspace and refer to companies by display_name. Never print an internal numeric tenant id, and never show a raw data_source_id — pass the opaque handle, show the name.

Only then does the work begin. Once the workspace, the data sources and their tools are confirmed, resolve every task against what was actually found: what is available now decides which steps are [auto], which are [gated] and which fall to [manual]. Where the confirmed tools cannot answer the request, say so and ask — do not substitute an assumption for a capability.

The catalogue is authoritative. Take exact tool_name values from the Gate 2 listing — names are not uniformly styled, some underscored, some hyphenated. Do not invent a tool name. On UNKNOWN_TOOL, read the valid names from the error and retry. Never call a tool whose effective_policy is disabled.

This map is built fresh every run and held only for this run. It is written out in the coverage sheet, never written back into this file.

Gate 0 — Confirm the workspace

Call list_workspaces with no arguments.

• One workspace → read the workspace name back and wait for an explicit yes.
• Two or more (selection_required) → ask in chat whether this is single- or multi-workspace, then which workspace(s) by name, then call again with workspace_ids=[…] and mode="single" / mode="multi".

Never auto-pick. Never print an internal numeric tenant id — name the workspace, pass the opaque ws_… handle.

When this skill is invoked as a handoff from another skill, the gates have usually run already in the same conversation. Confirm the entity in one line rather than repeating the sequence — but never build against an entity that was not confirmed.

Gate 1 — Discover live datasources and settle the entity scenario

Call get_agent_datasources with the confirmed workspace_id.

• connected: true → in scope.
• connected: false → excluded, and named as excluded. Surface any reconnect_url.

This gate answers the original's "Accounting system" input. The connected platform is known — so the import format in Step 7 is determinable rather than a question. Ask only when the platform's import template is itself the uncertainty.

Settle the entity scenario:

• Single-entity — ask which company by display_name; the entry is built against that chart.
• Multi-entity — ask which set, and which entity this entry belongs to, before building it. For an intercompany entry, both entities are in scope and both sides are built. See the cross-entity step.

Refer to companies by display_name; never show the raw data_source_id.

PART A — Explore the confirmed sources, and personalise this run

The workspace and its data sources are settled. This part finds out what they expose and which of it serves this request — the tool catalogue in Gate 2, then what is already known about this entity plus whatever still has to be asked in Gate 3. The result is a run shaped around these books, not a generic template.

Gate 2 — Discover enabled tools → build the capability map

Write tools are out of scope — always

get_datasource_tools describes what the connection could do. This skill uses only the reads.

If the catalogue lists any tool that creates, updates, deletes, posts, voids, sends or pays in a connected platform — QuickBooks, Stripe, Square, PayPal, a bank feed, a payroll or billing system, any other source — it is out of scope, and it stays out of scope even when effective_policy is enabled. A permission to write is not an instruction to write. Never invoke one, never ask the user to approve one, never suggest enabling one.

This holds for every connected platform, not only the books. Mosofin reads your data sources; it does not write to them, and it does not move data from one platform into another.

If a step appears to need a write, that step is [manual]. Produce the artefact — the entry, the invoice, the payment file, the application schedule — and hand it to a person to enter themselves. Say so plainly in the output, so nobody assumes it was done.

Hard stop — the four ways a write could slip through

Never route around this rule. Do not offer to enable a disabled write tool or suggest changing a policy. Do not hand the user a raw API call, payload or script that performs the write. Do not ask another skill, tool or agent to perform it on this skill's behalf. Do not defer it to a later step in the hope it becomes permitted.

There is no path through this skill that ends in changed data. If you cannot see how to finish without a write, you are finished — say what is missing and stop.

For each in-scope datasource (and per company file when several are live — pass data_source_id), call get_datasource_tools. Bucket every tool by effective_policy:

Resolve every validation in Part B against these buckets. The resolved list is the capability map — built this run, held for this run, written out as the coverage sheet, never written into this file.

If the chart read is disabled, the original's fallback applies in full — descriptive account names labelled "PLACEHOLDER — REMAP REQUIRED". Say which fallback is in use and why; do not invent numbers.

Rules that bite hardest here:

• Read the real tool name from the catalog, never from memory. Names are not uniformly styled — some underscored, some hyphenated.
• A near-substitute is not a substitute:
  • A validated entry is not a posted entry. Nothing here records anything.
  • An account existing is not an account being appropriate. The chart says it exists; judgment says it fits.
  • An open period in the system is not an open period for reporting. Statements may already be issued even where the software permits posting.
  • A balance in an accrual account is not proof the accrual covers this item — it is a prompt to check.
• There is no posting capability, no materiality policy and no tax rule engine on this surface.

Gate 3 — Profile the entity, then interview the user

Call the platform's company-profile tool (on QuickBooks, get_company_info) for each in-scope entity.

Derive silently what the profile answers: legal name, functional currency, fiscal calendar and year-end, country / region.

Ask the user what actually changes the work — the original Inputs table, minus the four the connected books now answer:

If no chart of accounts is available: ask the user for the specific account names and numbers they use. Do not invent. Acceptable fallback: descriptive account names with a clear "PLACEHOLDER — REMAP REQUIRED" label.

Ask as one short batch. Propose defaults where reasonable — but never default an account, a tax treatment, or a materiality threshold.

On later runs, read stored preferences first (Step 9), confirm in one line, and ask only what changed. The account map, the entry templates, the import format and the numbering convention persist; balances and validation results are re-obtained every time.

PART B — The domain work

Every step below is the original procedure, unchanged in count, order, or substance, with plain-language wording, an [auto] / [gated] / [manual] verdict, and the typical evidence tool added.

Never drop a task because no tool covers it. The event itself and the tax rules are legitimately [manual].

Tool names in italics are typical. Resolve real names and policies from your Gate 2 catalog.

Step 0 — Fetch the evidence (grounding) — Mosofin addition

Batch independent reads into one message — the chart, the balances and the validation populations do not depend on each other. Never serialize them.

The server is stateless: pass data_source_id on every call, including retries.

Typical batch, per in-scope entity:

• search_accounts — the chart with account numbers, names and types — usually [auto]. The single most important read here
• get_trial_balance / get_balance_sheet — current balances on the accounts being touched — usually [auto]
• get_general_ledger on the accrual, prepaid, deferred and clearing accounts — the Pitfall 1 check — usually [auto]
• search_invoices / search_bills — whether a document already exists for this item — usually [auto]
• search_journal_entries — the numbering convention in use, and whether the period still accepts entries — usually [auto]
• get_company_info — legal name, functional currency, year-end — usually [auto]

Handle the envelopes:

• approval_required → ask the user in chat, then re-invoke the same tool with approved=true.
• entity_required → ask by display_name, then pass that data_source_id.
• tool_policy_disabled → convert that validation to [manual] and record the gap. If it is the chart read, state that placeholder coding is in use.
• UNKNOWN_TOOL → read the valid names from the error; do not guess.
• Dead connection → surface the reconnect_url.

Check the mock flag. mock: true is fixture data — an entry validated against fixture accounts may reference accounts that do not exist in the real chart, and it would be imported.

Step 1 — Identify the JE type

State the JE type explicitly in the header.

Mosofin note on Closing: check whether the platform closes the year automatically before proposing a closing entry — see closing-entries-and-trial-balance. [auto]: P&L accounts at zero after year-end with no visible closing journal means the system did it, and a manual closing entry would double-close retained earnings.

Step 2 — Apply the accounting equation

Every JE must balance: Total Debits = Total Credits.

Normal balance rules:

Walk through substance:

1. What did the entity receive? → Debit (asset / expense) or Credit reduction
2. What did the entity give up? → Credit (asset reduced) or Debit reduction
3. What obligation was created or settled? → Credit (created) or Debit (settled)

[auto] enforcement: account types are readable, so the sign convention is checkable rather than assumed. A debit to a revenue account, or a credit to an expense account, is not wrong in itself — but it is unusual enough to warrant the memo saying why, and the check surfaces it.

Step 3 — Construct the entry

Standard format:

Date: YYYY-MM-DD
Entry #: [auto or user-supplied]
Type: [Standard / Accrual / Reclass / etc.]
Reference: [Invoice #, contract, supporting doc]
Memo: [One line describing the business event]

Account No. | Account Name | Debit | Credit | Line Memo
------------|--------------|-------|--------|----------
[from user's COA] | [from user's COA] | $X | | [purpose]
[from user's COA] | [from user's COA] | | $X | [purpose]

Total                                 | $X | $X | ✓ Balanced

Required for every entry:

• Posting date
• Entry type
• Header memo — business purpose in plain language
• ≥ 2 lines — one DR, one CR
• Line-level memos when multiple debits or credits
• Balance check at bottom

Mosofin additions to the header, since the entry travels beyond this conversation:

• PROPOSED ENTRY — NOT POSTED, stated plainly
• The entity by display_name
• The validation result from Step 6
• The source: which skill or analysis produced it, and the tool calls behind any figure

Account numbers and names come from the live chart (search_accounts, [auto]) — exactly as they appear there, including formatting, so the import matches.

Step 4 — Common JE patterns

DR is a debit, CR a credit. Every pattern below is a proposal.

Accrual (expense incurred, not yet invoiced)

DR  Expense account                       $X
    CR  Accrued Liabilities (BS)              $X
Memo: Accrue [service] for [period] — invoice expected [date]

Flag as reversing on first day of next period.

Prepaid expense (paid in advance)

At payment:
DR  Prepaid Expense (BS)                  $X
    CR  Cash / AP                             $X

Periodic amortization:
DR  Expense                               $X / periods
    CR  Prepaid Expense                       $X / periods

Suggest prepaid-amortization-schedule for the full schedule.

Deferred revenue

At receipt:
DR  Cash                                  $X
    CR  Deferred Revenue (BS)                 $X

As earned:
DR  Deferred Revenue                      $X / period
    CR  Revenue                               $X / period

Depreciation

DR  Depreciation Expense                  $X
    CR  Accumulated Depreciation              $X
Memo: Periodic depreciation — [asset class or specific asset]

Reclass between accounts

DR  Correct Account                       $X
    CR  Incorrect Account                     $X
Memo: Reclass [item] from [wrong] to [correct]; original entry #, date

[auto]: the original entry is findable (search_journal_entries, get_general_ledger), so the memo's entry number and date can be filled from the ledger rather than from memory.

Correcting prior error

Never delete or alter the original entry if posted. Either:

• Reverse and rebook — full reversal + correct entry — preferred for material errors or audit-trail needs
• Net correction — just the difference — for small true-ups

Reference the original entry # in the memo either way.

Mosofin note: Mosofin cannot delete or alter anything, which enforces the first sentence structurally. And the original entry is readable, so the reversal can be built from what was actually posted rather than from what someone remembers posting.

Intercompany

Always create mirror entries in both entities:

Entity A (lender):
DR  Intercompany Receivable — Entity B    $X
    CR  Cash                                  $X

Entity B (borrower):
DR  Cash                                  $X
    CR  Intercompany Payable — Entity A       $X

Flag for intercompany-reconciliation at period end.

[gated], and stronger here than the original assumes: where both entities are connected, both sides are built against their own real charts — the account names differ between entities and using the correct ones in each is the difference between an entry that imports and one that fails. Where the counterparty is not connected, build the near side and say the far side could not be built against a real chart.

Reversing entries

Date: first day of next period. Flip debits and credits from the original. Reference the original entry #. [auto]: the original is readable, so the flip is mechanical and the reference is exact.

Closing entries (year-end)

Close revenues and expenses to Income Summary — or directly to Retained Earnings if the system supports it. Verify net result ties to P&L net income / loss. [auto] to verify: get_profit_and_loss gives the figure the entry must produce.

Step 5 — Tax effects (jurisdiction-driven)

Only include tax lines if the user provided a jurisdiction and the transaction has tax implications.

For sales / output side:

• VAT / GST / sales tax collected → Liability (Tax Payable / Output Tax)
• Multi-jurisdiction sales → separate lines per jurisdiction
• Tax-exempt or zero-rated → no tax line; note the exemption reason

For purchases / input side:

• Recoverable input tax (where the jurisdiction allows it) → Asset (Input Tax Receivable)
• Non-recoverable input tax → included in expense

If the jurisdiction's rules aren't clear, ask the user. Do not assume.

[gated]: the tax accounts and any configured tax codes are [auto], so a tax line can name a real account and a real code — but which code applies remains [manual]. See gl-coding-assistant, where the same split appears.

Step 6 — Validation

Before delivering, verify — and here most of these are performed, not merely intended:

• ✅ Debits = Credits, to the penny — [auto]
• ✅ All accounts exist in the provided COA — [auto] against the live chart. An account that does not exist is a hard stop, not a warning
• ✅ Date is within an open accounting period (warn if not) — [gated]: entries existing after a date, or the platform's own period status, indicate it. Warn about reporting status too — a period the software will accept may already have been reported
• ✅ Memo explains the business purpose — not "JE" or "adjustment"
• ✅ References to source documents included — [gated]; findable where the document is in the system
• ✅ Reversing flag set correctly if applicable
• ✅ Sign conventions match account types — [auto] from account types
• ✅ No P&L accounts used for pure BS reclasses without a stated reason — [auto] from account types
• ✅ Tax lines present if jurisdiction + transaction warrant — [gated]

Mosofin additions to the validation set — the Pitfall checks, run as queries:

• ✅ No existing accrual already covers this item — [auto]. The Pitfall 1 double-count check. Read the relevant accrual account before proposing a payment or invoice entry
• ✅ No open invoice exists for a receipt being booked to revenue — [auto]. Pitfall 2
• ✅ A loan payment is split between interest and principal — [auto] support from the liability balance. Pitfall 5
• ✅ The target is not a control account — [auto]: manual entries to revenue, cash, AR or AP control accounts are auditor red flags, and whether an account is one is readable
• ✅ The entry does not exceed materiality without a review flag — [gated]; the threshold is [manual]
• ✅ A clearing or suspense account used here is expected to clear — [auto] to report its current balance and age

Report the validation result with the entry. A proposal that says "validated: 9 of 9 checks passed; account 6410 confirmed in chart; period open; no matching accrual found" is one a reviewer can post quickly. One that says "validated" is not.

Step 7 — Output format

Single JE → markdown table inline. Multiple JEs / import-ready file → .xlsx or .csv.

Generic columns:

| Date | Entry # | Account No. | Account Name | Debit | Credit | Memo | Reference | Class/Dept | Currency |

System-specific column headers — and the connected platform is known from Gate 1, so use its format by default rather than asking:

• QuickBooks Online: JournalNo, JournalDate, Line Description, Account, Debit, Credit
• Xero: *Narration, *Date, Description, *AccountCode, *TaxRate, *Amount
• NetSuite: External ID, Date, Account, Debit, Credit, Memo, Subsidiary, Department, Class
• Sage Intacct / 50 / 200: format varies — ask the user for the import template
• Microsoft Dynamics 365 / Business Central: Journal Batch, Posting Date, Document No., Account Type, Account No.,

…(truncated)