BD statement-extract-and-prove
Converts bank, credit card and brokerage statement PDFs, invoices and vendor price lists into clean CSV or Excel, then proves the extraction is complete and correct - opening balance plus transactions equals closing balance, every running balance chains, page totals, row counts and summary totals all tie, signs and decimal separators are consistent. Handles digital PDFs with word-position column bands and detects scanned PDFs that need OCR first. Use it whenever the user wants to convert a bank statement PDF to Excel or CSV, extract transactions from a statement, turn a PDF table into a spreadsheet, import statements into QuickBooks or Xero, or says the converted statement does not balance, rows are missing, or signs look wrong, even if they do not ask for a proof. Never reads amounts off the page image when a text layer exists, and never fixes a failed tie-out by guessing.
Converts bank, credit card and brokerage statement PDFs, invoices and vendor price lists into clean CSV or Excel, then proves the extraction is complete and…
As a process D 39/100 · Unfinished process — References files that are not bundled: scripts/reconcile.py
How to improve
- The text references files that are not there: add them or drop the references.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
missing-refreference to a missing file: scripts/reconcile.py
Process rating: all ten parameters 39/100
- 0Tools and files. 1 referenced file(s) missing: scripts/reconcile.py
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 3 mutating operations with no state check
- 70When it triggers. States when to use, but not when not to
- 100Steps. 17 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2007 tokens
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 886: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 12 headings
- +3Step-by-step instructions: 17 items
- +4Has examples (5 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 82.