agentsclimarketplace

Report output validator

Skill nicholashidalgo/claude-skillforge/data-engineering/report-output-validator

Install
npx -y skills add nicholashidalgo/claude-skillforge --skill report-output-validator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Validate report outputs with Python using row-count checks, subtotal tests, duplicate-grain checks, and comparison summaries.

SKILL.md

1.8 KB, as published. Nobody here has run it

Runtime Configuration

version: "1.0.0"
gotcha_pack: "sql-data-gotcha-pack"
gotcha_pack_version: "1.0.0"
gotcha_enforcement: "block_on_high"

Purpose

Validate report output before release or after logic changes.

Checks

  • row counts
  • duplicate grain checks
  • subtotal to total tie-out
  • period coverage
  • category coverage
  • comparison to prior or source extract
  • reasonableness checks

Output format

  1. Validation scope
  2. Pass or fail checks
  3. Exceptions
  4. Python script
  5. Release risk summary

Gotcha Enforcement

Every validation script produced must satisfy these rules. HIGH violations block output. MEDIUM violations appear in Validation results as warnings with explanation.

IDSevCheck
G002HIGHIf the report involves a join, state the expected cardinality and test for fan-out
G003HIGHEvery aggregation in validation code documents its NA/null behavior explicitly
G007HIGHValidation logic must not replicate the report's own SQL or transformation logic
G009MEDIUMNull rate for every critical measure column must be checked and surfaced
G012HIGHConfirm grain, period, and filter alignment before comparing source to report
G014HIGHAt least one measure value must be reconciled; row count alone is not sufficient
G015MEDIUMIf net difference is zero, run a segment-level check before calling the report clean

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.