Find out which clients and jobs actually make money -- allocate real costs (labor hours, materials, subs, overhead) against what each job billed, and rank every client by true margin. Most service businesses discover their biggest client is their least profitable.
SkillScore73Maintained
Automatic
◆Finance & Accounting pack
Last verified 8/24/2026
What it changes
Passed 16 of 22 test cases with the skill, and 15 of 22 without it.
Without skill68%
With skill73%
Checks it passed that the model alone missed
Rejects using unburdened base wages ($5000) as the complete labor expense.
Highlights any rank reversal between Job 1 and Job 2 resulting from the different allocation methods.
Recommends tracking performance across multiple future engagements before making a termination decision.
and 1 more in the full report
3 fixed13 passed either way4 still failing2 broke
Pass-rate lift
+5 pp
Tokens
+45%
Agent turns
0%
Response time
+17%
Model gemini-3.6-flashMeasured 8/24/2026v116 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 3
Prompt
We paid our senior developer $50/hr on paper for 100 hours on Project Alpha, which billed $8,000 in revenue with $1,000 in direct software costs. Calculate our labor expense and gross profit for this project.
What was checked, without and with the skill
Checked
Without skill
With skill
Multiplies the base wage rate of $50 by a labor burden factor between 1.25 and 1.40 to calculate total labor cost.
Missed
Passed
Rejects using unburdened base wages ($5000) as the complete labor expense.
Missed
Passed
case-02 · gemini-3.6-flash · 8/24/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md27 linessingle file
Job Profitability Analyzer
Revenue is vanity, margin per job is survival. Service businesses (agencies, contractors, consultancies, trades) routinely lose money on their "best" clients because nobody allocates the hours. Input: invoices/revenue by job or client, time tracking exports, materials/sub costs, payroll rates, and a rough overhead number. Output: profit per job, per client, per service line -- with receipts.
Workflow
Build the job ledger. Match every revenue line to a job/client. Match every cost: labor (hours x loaded rate -- wages plus taxes/benefits, ~1.25-1.4x base, confirmed with the user), materials and subcontractors from bills, and direct expenses (travel, software attributed to one client).
Allocate overhead. Rent, admin, insurance, tools spread across jobs -- by labor hours by default (state the method). Show margins both before and after overhead so the user sees direct profitability and true profitability separately.
Rank.profitability-report.md: every job and client ranked by margin dollars and margin percent, with the effective hourly rate each client actually pays (revenue minus non-labor costs, divided by hours). This one column reorders most people's client list.
Diagnose the losers. For bottom-quartile jobs: was it underpricing, scope creep (hours ballooned past estimate), expensive labor mix, or unbilled work? Cite the timesheet and invoice evidence per diagnosis.
Prescribe. Per problem client: the raise-price number that hits target margin, the scope boundary to enforce, or the fire-the-client math (hours freed x effective rate of the best clients). Pair with pricing-strategy for the repricing conversation.
Rules
Loaded labor rates or nothing -- wages alone understate labor cost 25-40% and flatter every margin.
Unbilled hours are the finding, not noise: track estimated vs. actual vs. billed hours per job explicitly.
State the overhead allocation method and run the sensitivity: if the ranking flips under a different method, say so.
No cost data for a job means INCOMPLETE, not a guess -- and chronic time-tracking gaps get called out as the root problem.
Small samples get honesty: one bad job does not condemn a client; three do. Note the n per client.
Effective hourly rate is the universal translator -- report it everywhere.
Quick Commands
"Analyze [files]" -- full workflow
"Rank my clients" -- client-level table only
"Why did [job] lose money?" -- single-job autopsy
"What should I charge [client]?" -- the repricing math