Back to results
Explore the library298 guides total · goals, work & tasks
By goal

Choose what you want to achieve.

By work or interest

Choose a setting you work or spend time in.

By task

Choose the kind of help you want.

Counts show all available guides. Filters combine to narrow your results.

Create work & incomeGuide

Check a small craft order’s costs before quoting

Add supplied materials, labor and batch costs, then test the first whole-penny price that meets a fee-aware residual target.

Illustrative resourceNot tool-testedPeople checking their own supplied small-order cost arithmetic before making a separate quote decision
What you’re working towards

A reusable blank cost worksheet and exact worked breakdown, with a checked whole-penny price boundary under the supplied fee and residual rule. Missing real costs remain unknown, and choosing or sending an actual quote stays with the owner.

Download playbook .md

Your choice also saves this guide’s current worked-example checkmarks. Only this browser; no account or sync. View saved guides

Saving needs JavaScript. Reading and downloading this guide does not save it.

Included example · illustrative MD file

Download the reusable cost worksheet

Blank editable Markdown worksheet with exact formulas, a penny search, explicit unknowns and separate owner decision fields. It does not recalculate automatically.

Not tool-tested. Review and adapt before using it for a real project. Your browser may download this file when you choose View example file.

Source files to work with

Make a start

One step at a time.

5 steps

Start with the synthetic worked example or other non-sensitive inputs. Run one prompt at a time in your chosen AI tool. Replace any bracketed placeholders with the requested material, check each reply, and supply the source or earlier outputs that step needs. Paste them again if the tool lacks that context.

Set up the worked example

Open

Open a new text-AI chat, a plain-text editor and a calculator that can keep exact fractions or decimal amounts. Download craft-cost-worksheet.md and save your own copy as my-craft-worksheet.md. Markdown (.md) is editable text; this worksheet and the CSV do not calculate when edited.

Add the sample

Open craft-cost-source.md, or copy all of “The starting point” (Synthetic input), and replace step 1’s [paste complete source] with it. Keep the source unchanged. Send later prompts one at a time in the same chat. Step 2 needs that complete checked source. For step 3 paste the source, your checked cost breakdown and these sample candidate prices: GBP 60.27 and GBP 60.28. Step 4 needs the complete source and checked supplied-cost total, not just a proposed price. Copy the costs, fee comparisons, minimum-search evidence and unknowns into my-craft-worksheet.md; save and reopen it before step 5. In step 5 paste the original source and the reopened worksheet in their separate fields. “The intended result” (Illustrative output), worked explanation and answer JSON are comparison answers. For the optional advanced sample exercise supply the entire original source; explicitly change quantity to 13 and labor to a newly supplied 100 minutes for those 13, retaining GBP 12/hour; explicitly confirm all per-item, whole-order, fee/target inputs and unknowns unchanged. A new chat needs the whole source and revised-input statement again.

Start with the synthetic example, then adapt each prompt to your task.

01

Copy the blank worksheet and keep the source separate

Start with one order quantity. Label each amount as per item, whole order or labor for this quantity. In the example, paper, cover and thread repeat 12 times; the box, filler and setup apply once. Missing real costs stay visible.

Your starting prompt
Organize this supplied craft-order input into a cost worksheet without choosing a price yet. Preserve the quantity, currency, source IDs and every exact amount. Separate per-item costs, whole-order costs, labor, fee rules and the owner’s residual target. Mark missing values unknown, never zero. Validate that quantity is a positive integer and monetary values are nonnegative whole-penny amounts. Require labor explicitly supplied for this exact quantity. Hold missing or invalid required inputs; do not infer tax, shipping, overhead, provider fees or a market price. Keep the original source unchanged.

SOURCE:
[paste complete source]
Expect

A worksheet separating three per-notebook cost lines, three whole-order lines, quantity-specific labor, the fee and residual rule, and unknown real costs.

Check

The order is 12 notebooks in GBP. M01/M02/M03 repeat per item; B01/B02/B03 apply once. L01 is 90 minutes at GBP 12/hour for exactly 12 notebooks. Tax, shipping, overhead and other real costs remain unknown.

If it differs

If a basis, quantity or required labor input is missing, compare with the original source and hold the calculation. A supplied zero may be valid; a blank is not zero. Keep real-cost completeness and any quote decision pending.

02

Add the supplied costs once, using the right basis

Materials total £19.20. The supplied 90 minutes at £12 per hour adds £18.00, and the three whole-order lines add £7.00. Their £44.20 sum is a supplied-cost total; it does not show that all real costs are covered.

Your starting prompt
From the validated source, show each per-item multiplication by quantity, each whole-order cost once, and labor as supplied minutes × hourly pennies ÷ 60. Use integer pennies and exact fractions, not binary floating-point rounding. Preserve every source ID exactly once. If labor gives fractional pennies without a supplied rounding rule or whole-penny total, hold. Show materials per item, material total, whole-order total, labor and combined supplied-cost total. Keep all unlisted real costs unknown. Do not scale labor if quantity changes.

VALIDATED SOURCE:
[paste]
Expect

Seven source-linked cost lines and a supplied-cost subtotal, with exact labor arithmetic.

Check

Check 12 × (85 + 60 + 15) = 1,920 p; 200 + 100 + 400 = 700 p; 90 × 1,200 ÷ 60 = 1,800 p. Total: 4,420 p, or GBP 44.20. Setup is a money line, not extra labor minutes.

If it differs

If the subtotal differs, check each ID and its basis once; ask for corrected line totals and the full sum. If new inputs produce fractional-penny labor, require an owner-supplied rounding rule or whole-penny total before continuing. Never scale this labor to a different quantity.

03

Apply the fee and compare the target exactly

The fictional rule rounds 5% of the whole order price up to a penny and adds £1 once. The target is 20% of customer price. At £60.27, £12.05 residual is below its exact £12.054 target, even though rounding the target down would hide that failure.

Your starting prompt
Use only the supplied fee rule and target. Let P be the positive whole-penny order price, C the listed cost total, r the exact variable rate, f the fixed fee and m the exact residual target rate. Compute V = ceiling(r × P), then residual R = P − C − V − f. Compare R with the exact m × P without rounding the target. If m = a/b, use b × R ≥ a × P. An exact whole-penny variable fee stays unchanged. Hold r + m ≥ 1, negative values, missing rules, other fee bases or unsupported rounding. Do not substitute a markup or apply the fixed fee per notebook.

SOURCE, COSTS AND CANDIDATE PRICES:
[paste]
Expect

A candidate table separating the rounded-up variable fee, fixed fee, exact residual target and residual before unlisted real costs.

Check

At 6,027 p and 6,028 p the 5% fee rounds up to 302 p; add 100 p once. Residuals are 1,205 p and 1,206 p. Their exact targets are 1,205.4 p and 1,205.6 p. The first fails, the second passes. “Ceiling” rounds a fractional penny upward; a whole penny stays unchanged.

If it differs

If the target is rounded to force a pass, keep the fractional penny or compare 5 × residual with price using integers. If the fixed fee is per item, restore once per order. A different fee basis/rounding needs a new model; do not substitute a familiar provider’s rule.

04

Prove the first qualifying whole penny

The continuous lower bound is £60.2666… before the fee rounding. The only candidates from the necessary lower bound through a sufficient upper bound are £60.27 and £60.28; the second qualifies. The local integer checker also enumerated every positive penny through £60.28.

Your starting prompt
Find the smallest qualifying positive whole-penny price using the exact supplied model. Let d = 1 − r − m, require d > 0, and set L = max(1, ceiling((C + f)/d)), U = max(1, ceiling((C + f + 1)/d)), all in pennies. The extra 1 is one penny. Explain why prices below L fail and U is sufficient because upward fee rounding adds less than one penny. Test every integer P from L to U in increasing order and show the first qualifying result plus the preceding penny with exact targets. A predecessor test alone does not prove the minimum in general. Mark minimum unverified if the exact search is not performed. Do not call the result an actual quote or predict customer acceptance.

VALIDATED SOURCE AND SUPPLIED-COST TOTAL:
[paste]
Expect

A proof that 6,028 p is the first qualifying positive whole-penny price, including the bounds and exact tests. Save the complete calculation in my-craft-worksheet.md.

Check

The necessary lower bound is 6,027 p and sufficient upper bound 6,028 p. Independently test both using the prior step. The extra 1 in the upper-bound formula means one penny. Checking only a passing price and its predecessor is not a general minimum proof.

If it differs

If no exact interval search was actually performed, mark the minimum unverified and retain the calculated candidate only. Do not copy a prior checker’s run claim. Resolve missing rules or a rate-plus-target sum at/above 100% before searching.

05

Audit the result and leave the owner’s decision open

Use the independent cost breakdown and exact comparison to check the AI’s answer. The source, blank worksheet and worked answer are separate downloads. The owner still needs real cost and fee checks before choosing any quote.

Your starting prompt
Audit this result against the unchanged complete source. Recalculate the seven cost lines, fee rounding, fixed fee, exact target and whole-penny minimum independently. Reconcile price = listed costs + variable fee + fixed fee + residual. Test the immediately lower penny, missing labor, a changed quantity with old labor, negative values and r + m at or above 100%. Preserve missing tax, shipping, overhead and other real costs as unknown; do not silently add zero. Label the residual as before any unlisted real costs. Separate arithmetic and file checks from unperformed AI-provider, real-quote, human-use or customer-acceptance checks. Leave the actual price decision to the owner. Do not forecast sales or profit, send a quote, or start checkout.

SOURCE:
[paste]

CANDIDATE WORKSHEET:
[paste]
Expect

An audit of the reopened worksheet, plus separate invalid-input tests and unresolved real-use questions.

Check

Reconcile 4,420 + 302 + 100 + 1,206 = 6,028 p. On separate source copies test missing labor, quantity 13 with labor still for 12, a negative cost and target 95% with variable fee 5%; each requires a hold. Check the 6,027 p failure without rounding its target.

If it differs

Keep a failed worksheet, correct a new revision, save/reopen and repeat the affected checks. Report tests not run honestly. Actual costs, real fees, tax treatment and the owner’s price choice remain separate; this arithmetic boundary is not a sent quote, earnings forecast or evidence of customer acceptance.

Optional: improve the result

Compare an explicitly supplied revised order

Keep the original source and create a new scenario. For practice, 13 notebooks with newly supplied 100 labor minutes at £12/hour produces £47.80 listed costs and a £65.08 arithmetic boundary under the same fictional fee and target. The 100 minutes is a new input, not an inferred productivity rule.

Advanced prompt
Create a separate revised scenario from the original source using only the explicit changes supplied below. If quantity changes, require new labor supplied for that exact quantity before calculating; do not scale old labor. Keep unchanged per-item amounts and whole-order amounts only if the owner explicitly confirms they still apply. Validate all inputs, recalculate every cost and enumerate the exact whole-penny fee-aware boundary. Show original and revised quantities, supplied labor, listed costs and first qualifying prices. Preserve all unknown real costs and record no actual quote decision. A difference is a change in the supplied arithmetic model, not measured savings or forecast earnings.

ORIGINAL SOURCE:
[paste]

EXPLICIT REVISED INPUTS AND CONFIRMED UNCHANGED AMOUNTS:
[paste]
Optional: use your own input

Check your own supplied small-order costs

Use an anonymized order and non-sensitive cost data you are permitted to share; no customer details are needed. Copy craft-cost-worksheet.md to my-craft-worksheet.md. This route uses one currency with a hundred minor units per major unit, whole-minor-unit prices, one upward-rounded percentage fee on the whole order price plus one fixed fee. A different fee/currency model needs separate formulas. Start a fresh chat with the prompt below and check the cost bases before continuing.
“Using MY CHECKED SOURCE, calculate each per-item cost × my quantity, each whole-order line once and labor for this exact quantity. Use integer minor units and exact fractions. For time-based labor use minutes × hourly minor units ÷ 60; apply only an explicitly supplied rounding rule or total, otherwise hold a fractional result. Show every source ID once and all subtotals. Keep unlisted real costs unknown; do not scale labor. MY CHECKED SOURCE: [paste complete validated inputs].”
Independently check each line and sum; repair errors before sending:
“Apply only MY FEE AND TARGET RULES to MY CANDIDATE ORDER PRICES. In integer minor units let P be price, C my checked listed costs, r the exact variable rate, f the once-only fixed fee and m the exact residual percentage of price. Calculate V = ceiling(r × P), R = P − C − V − f and compare R with exact m × P without rounding that target. Hold negative/missing inputs, unsupported rounding or fee bases, or r + m ≥ 1. SOURCE RULES, CHECKED COSTS AND MY EXPLICIT CANDIDATE PRICES: [paste].”
Choose positive whole-minor-unit practice candidates yourself; these are calculation tests, not chosen quotes. Check fee rounding and the exact comparison, then send:
“Find the minimum under MY COMPLETE VALIDATED MODEL. Require d = 1 − r − m > 0. In minor units calculate L = max(1, ceiling((C + f)/d)) and U = max(1, ceiling((C + f + 1)/d)); the extra 1 is one minor unit. Explain the bounds and test every integer price from L through U in order with exact fee/target comparisons. Show the first qualifier and, if positive, its predecessor. A predecessor alone is not proof; if the full exact search is not performed label minimum unverified. Do not decide an actual quote. COMPLETE MODEL AND CHECKED COST TOTAL: [paste].”
Independently verify the bound and every candidate in that interval; if too large to check with your available tools, keep the minimum unverified. Save all source-linked calculations, search evidence and unknowns in my-craft-worksheet.md and reopen it.
“Audit MY REOPENED WORKSHEET against MY ORIGINAL SOURCE and MY INDEPENDENT CHECKS. Reconcile price = listed costs + variable fee + fixed fee + residual. Check every cost basis, quantity-specific labor, fee rounding, exact target and minimum proof. On separate copies test a missing required value, revised quantity with old labor and an invalid rate/target sum. Preserve unknown real costs and any unverified minimum. Do not choose tax treatment, forecast earnings, send a quote or start checkout. ORIGINAL SOURCE, REOPENED WORKSHEET AND CHECK RESULTS: [paste].”
Keep the old version; correct, save/reopen and recheck any failed result. Every count, cost, rate and candidate comes from your own input, not the sample. The original advanced exercise remains sample-scoped. An own-input revision requires a separate complete source, explicit changes, newly supplied labor for any changed quantity and confirmation that unchanged amounts still apply. The owner reviews real cost completeness, actual fees and tax treatment before separately choosing any quote.

Your own-input prompt
Help me check ordinary supplied small-order arithmetic, not make a quote or financial decision. First organize my input into per-item costs, whole-order-once costs, labor supplied for this exact quantity, one whole-order fee and my residual target as a percentage of price. Preserve source IDs and exact values; stop for my review before calculating. Require a positive integer quantity, one currency with 100 minor units per major unit, nonnegative whole-minor-unit monetary inputs, explicit labor quantity/basis, and complete fee/rounding rules. For fractional-minor-unit labor require my explicit rounding rule or whole-minor-unit total. This model supports only one percentage fee rounded upward to a minor unit plus a fixed fee once, with nonnegative variable and target rates whose sum is below 100%. Hold other bases, tiers, missing rules and contradictions. Explicit zero differs from missing. Do not assume tax, shipping, overhead or other real costs are zero. Later we will calculate, test candidates, prove a minimum if possible and audit a saved worksheet. Do not send, sell, charge, predict demand or decide tax treatment.

MY COMPLETE ORDER AND COST SOURCE
[Give order/version, item, positive quantity, currency/minor unit; each cost ID, amount, per-item or whole-order basis and source; labor explicitly for this quantity, minutes/hourly amount or supplied total and any rounding rule; complete fee basis, rate, fixed amount and rounding; your residual percentage of customer price. List unknown real costs, confirmations and who reviews the result.]

Check your own result

  • Your quantity, each source cost basis and labor quantity agree. Changed quantity requires newly supplied labor, not automatic scaling.
  • Your exact fee basis/rounding is supported; the fixed fee applies once, and the residual target is a percentage of price rather than a markup on costs.
  • Fractional-minor-unit targets are preserved; the independently checked interval search, not just a predecessor, supports any minimum claim.
  • All saved line totals reconcile. Explicit zero remains distinct from missing; unlisted real costs stay unknown and the residual remains before those costs.
  • Owner review of real costs, fees, tax treatment and desired price stays separate. No quote, earnings, acceptance or sale is established.
Show, then try

See the shape of the result.

This input is synthetic. The output is a written illustration of the intended result, not a verified AI-tool run.

The starting point

Synthetic input
# Craft order cost exercise

Complete original synthetic input, craft-cost-v1. The fictional order is 12 notebooks. Currency is GBP. These are owner-supplied teaching values, not actual supplier prices, a platform fee or a business quote.

## Costs supplied for the exercise

| ID | Item | Supplied amount | Applies to |
| --- | --- | ---: | --- |
| M01 | Paper | £0.85 | Each notebook |
| M02 | Cover | £0.60 | Each notebook |
| M03 | Thread | £0.15 | Each notebook |
| B01 | Box | £2.00 | Whole order once |
| B02 | Packing filler | £1.00 | Whole order once |
| B03 | Setup | £4.00 | Whole order once |
| L01 | Labor | 90 minutes at £12.00 per hour | Exactly 12 notebooks |

Labor is explicitly supplied for this quantity only. If quantity changes, ask for newly supplied labor for the revised quantity; do not multiply the old time by a quantity ratio. Do not treat the £4.00 setup line as extra labor minutes.

## Supplied fictional fee and target

- Apply one processing fee to the whole customer price for the order.
- The variable part is 5% of that price, rounded UP to a whole penny. A value already at a whole penny stays there.
- Add a fixed £1.00 fee once, after calculating that variable fee.
- The owner-supplied target is a residual after the listed costs and this fee of at least 20% of the customer price.
- Customer prices must be positive whole pennies. Find the smallest qualifying price under these supplied rules.
- Keep exact arithmetic for the comparison. Do not round the target down to two decimal places.

This is a percentage of customer price, not a percentage added to costs. Do not substitute a markup rule. Do not split the fixed fee across several assumed transactions.

## Missing real-world facts

Tax: unknown. Shipping: unknown. Overhead: unknown. Other real costs: unknown. Real cost completeness is not established. No actual tax treatment, platform fee, market price or willingness to pay is supplied. These unknowns are outside this fictional arithmetic total; they are not numeric zero or evidence that the real order has no additional costs.

## Required result and hold rules

Preserve each source ID and show per-notebook materials, whole-order material totals, labor, whole-order listed costs, the combined supplied-cost total, both fee components, the exact target and the residual before any unlisted real costs. Prove the minimum across whole-penny prices, including the penny immediately below the answer.

Hold the calculation when quantity is missing or is not a positive integer, when required costs/labor/rates are missing, when amounts are negative, or when the variable rate plus the residual target is at least 100%. Explicitly supplied zero costs are different from missing costs. If labor arithmetic yields a fraction of a penny, hold this worksheet until the owner supplies a penny rounding rule or a whole-penny labor total; do not invent rounding.

A qualifying result is an arithmetic boundary for the supplied model. The owner must decide whether and what to quote after checking real costs, real fees, applicable tax treatment and their pricing choice. No sending, checkout, sale or forecast is part of this exercise.

Copies text only. For image or file tasks, use the source files named in the steps too.

Download source .txt

The intended result

Illustrative output
# Worked craft order cost check

For this fictional order, the supplied-cost total is £44.20. The smallest whole-penny order price meeting the supplied fee and residual rule is £60.28. This is an arithmetic boundary before any unlisted real costs, not an actual quote or a prediction of what customers will pay.

## The listed costs

| ID | Calculation | Whole-order amount |
| --- | --- | ---: |
| M01 paper | 12 × £0.85 | £10.20 |
| M02 cover | 12 × £0.60 | £7.20 |
| M03 thread | 12 × £0.15 | £1.80 |
| B01 box | Once for the order | £2.00 |
| B02 packing filler | Once for the order | £1.00 |
| B03 setup | Once for the order | £4.00 |
| L01 labor | 90 ÷ 60 × £12.00 | £18.00 |

Materials per notebook: £0.85 + £0.60 + £0.15 = £1.60.
Materials for all 12: £10.20 + £7.20 + £1.80 = £19.20.
Other whole-order listed costs: £2.00 + £1.00 + £4.00 = £7.00.
Supplied labor for exactly 12 notebooks: £18.00.
Combined supplied-cost total: £19.20 + £7.00 + £18.00 = £44.20, or 4,420 pence.

The labor input belongs to this 12-notebook order. A different quantity requires a newly supplied labor input. Do not scale the 90 minutes automatically.

## Fee-aware price boundary

All calculations below use whole pennies or exact fractions. The variable fee is ceiling(5% × order price), then £1.00 is added once. The target is 20% of customer price, not a markup on cost.

| Candidate order price | Exact 5% before fee rounding | Rounded-up variable fee | Fixed fee | Residual before unlisted real costs | Exact 20% target | Result |
| --- | --- | --- | --- | --- | --- | --- |
| £60.27 | £3.0135 | £3.02 | £1.00 | £12.05 | £12.054 | Fails |
| £60.28 | £3.014 | £3.02 | £1.00 | £12.06 | £12.056 | Qualifies |

At £60.28: £60.28 − £44.20 − £3.02 − £1.00 = £12.06.
Total fee: £3.02 + £1.00 = £4.02. Costs plus fee plus residual reconcile to £60.28.

The exact target in pennies is 6,028/5 = 1,205.6 p. The residual is 1,206 p, so 5 × 1,206 = 6,030 ≥ 6,028. It exceeds the exact target by 0.4 p (£0.004). The residual fraction is exactly 603/3014 of the price, approximately 20.0066%; the exact inequality is the deciding test.

At £60.27 the residual is 1,205 p. Its target is 6,027/5 = 1,205.4 p, so 5 × 1,205 = 6,025 < 6,027. It misses by 0.4 p (£0.004). Rounding £12.054 down to £12.05 before comparing would incorrectly pass that price.

## Why this is the first qualifying penny

Ignoring upward fee rounding gives a necessary lower bound:

(4,420 p + 100 p) ÷ (1 − 1/20 − 1/5) = 4,520 ÷ (3/4) = 18,080/3 p = 6,026 2/3 p.

Therefore no whole-penny price below 6,027 p can qualify. A sufficient upper bound is (4,420 + 100 + 1) ÷ (3/4) = 6,028 p. Testing both whole-penny candidates in that interval rejects 6,027 and accepts 6,028. Separately, the local checker enumerates all 6,028 positive whole-penny prices from 1 through 6,028 using integer arithmetic and finds the same first result.

## What this does and does not establish

The arithmetic checks the supplied fictional model. Tax, shipping, overhead and other real costs remain unknown, not zero. Actual cost completeness, real provider fees and market acceptability are unverified. Before any real quote, the owner must validate the real inputs and decide the price separately. This result does not forecast sales or profit. No quote was sent, no checkout occurred, and no consumer AI workflow or customer acceptance was observed. Time savings are unmeasured.

The cost-breakdown CSV and answer JSON are static worked answers. Editing them does not recalculate anything. Copy the blank worksheet for a new scenario and retain the original source alongside the new supplied values.
Download result .txt
You’re the final check

Check the worked example.

These checks use the worked example and its particular facts. With your own material, check your actual names, numbers, sources and rules; do not reuse the sample’s values as the answer.

Unsaved checks stay in this tab and reset when you reload. If you choose Save guide, its current example checkmarks are kept in this browser when storage is available.

A few things to watch for.

  • Multiplying whole-order costs by quantity, or forgetting to multiply per-notebook materials.
  • Treating labor for twelve notebooks as automatically scalable to a different quantity.
  • Rounding the variable fee down or applying the fixed fee once per item instead of once per order.
  • Adding 20% to costs when the stated target is residual as 20% of customer price.
  • Rounding a fractional-penny target down before comparing, causing £60.27 to appear to pass.
  • Assuming a single continuous formula proves the smallest whole-penny result after fee rounding.
  • Filling unknown real costs with zero or presenting the illustrative boundary as a market-acceptable quote.
  • Editing a static CSV or answer file and assuming it recalculates automatically.

What has been checked?

Documentation
No dated source review recorded.
Included files
Scoped checks passed on 2026-10-14. Fixture checks; File parsing. See scope below.
AI workflow
Not run in a recorded consumer-AI workflow.
Time savings
Time savings not measured
Read the scope and evidence

craft-cost-worksheet.md

2026-10-14 · semantic fixture · passed

Literal worksheet structure and exact mathematical method checked, including the necessary/sufficient penny bounds and explicit missing-cost/labor holds. No human worksheet completion or rendered usability test.

  • Blank source/quantity, cost, labor, fee, exact target, unknown-cost and owner-decision fields retained
  • Penny bounds agree with independently enumerated fixture; no target rounding
  • Unresolved real costs stay unknown; quantity changes require new labor

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: 6be726e4375a7a81bae4e552a362d95ea9d1ba58ba5e3245582a7e46d9f5534d

craft-cost-worked.md

2026-10-14 · semantic fixture · passed

Exact worked text checked against source controls, integer-only full price enumeration and rational arithmetic. Real completeness and customer acceptance remain unknown.

  • Listed totals 1920/1800/700 p sum to 4420 p
  • 6027 p fails exact target and 6028 p qualifies first
  • Variable fee 302 p plus fixed fee 100 p; residual 1206 p before unlisted real costs

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: 6513cd6c0c8aace20da9f4171fd4db838b5cc2790238c3e2c53b4e8491bd5ef8

craft-cost-source.md

2026-10-14 · semantic fixture · passed

Original human-readable source checked for all supplied amounts/rules and no solved totals/price. No external provider fact check.

  • 12 notebooks, seven cost inputs, 5% upward-penny fee plus GBP 1, target 20% of price
  • Labor explicitly supplied for twelve only
  • No answer leaks; tax/shipping/overhead/other real costs are unknown

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: 58d4324df3d3668d4b5432800c498482b850b5b0cf06620bd44101f772abfd2a

craft-cost-source.json

2026-10-14 · semantic fixture · passed

Independent literal controls verify the complete synthetic source. Exact-decimal validation and in-memory invalid-input tests check the bounded arithmetic model.

  • 56 invalid source/price cases rejected, including missing labor, changed quantity with old labor and rate plus margin at/above 100%
  • Eight changed-input positive cases preserve exact arithmetic and distinguish explicit zero from missing
  • No floats substituted for source money/rates; source unchanged after checks

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: da25795ea3e3d0d85ffcf09ce6686f29bc4c36fb05e43969ec341ff6bd22d2cf

craft-cost-breakdown.csv

2026-10-14 · parser · passed

UTF-8 CSV parsing, exact seven-row/field identity, integer-penny cost arithmetic, source IDs and formula-trigger prefix inspection. No spreadsheet-app import tested.

  • Seven unique source IDs and explicit bases
  • Line totals 1020/720/180/200/100/400/1800 p sum to 4420 p
  • Labor minutes times hourly pennies divided by 60; static data only

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: 2f5cc5fe96f993a2961a896d822f014878aea0f2103f492ce214e5406721e4b1

craft-cost-answer.json

2026-10-14 · semantic fixture · passed

Separate exact-fraction recomputation and integer-only enumeration of all prices from 1 through 6028 p; invalid answer mutations are rejected. No actual quote decision.

  • First qualifying price 6028 p; exact target 6028/5 p and residual share 603/3014
  • Fifteen corrupted answer cases rejected, including down-rounded fee, omitted fixed fee, altered unknown and false quote-ready state
  • Whole-penny boundary and both fee rounding jump cases checked

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: c3c169638fdb59d191cca2d839c4a7c1fe8dea455b17b2dde16ac728c87040d1

craft-cost-assumptions.md

2026-10-14 · semantic fixture · passed

Literal role, scope and reuse-card checks against the complete source and tested hold cases. This establishes text presence and consistency, not human comprehension.

  • File roles distinguish source, blank and static answer
  • Missing labor, revised quantity and rate-plus-target holds included
  • Provider access, consumer AI, human use and actual quote remain unverified; savings unmeasured

Environment: Python 3.12.14 standard library on local Linux; read-only check-fixture.py with exact fractions, an independent integer-only oracle, JSON/CSV parsing and in-memory mutations. Captured verification-report.json. No network, consumer AI or real quote.

SHA-256: e5589adf3e0dd9ac1cb5dd6362e544ea81f5237f5d9e39d57b049774fe1bf19b

Copy these instructions

Your browser blocked automatic copying. Select the text below and copy it.