customer-import-audit

Audits database state after importing a new customer's spec workbook.

Updated Jun 15, 2026
One-click install
npx skills add https://github.com/ravenslight2010/Production-run-calculator --skill customer-import-audit-ravenslight2010
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: customer-import-audit
Source: https://github.com/ravenslight2010/Production-run-calculator/tree/main/.agents/skills/customer-import-audit
Command: npx skills add https://github.com/ravenslight2010/Production-run-calculator --skill customer-import-audit-ravenslight2010

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After importing a new customer's spec workbook, teams need confidence that profiles, recipes, aliases, pools, and customer tags landed correctly without corrupting existing data. This Skill provides a read-only, evidence-based audit procedure that compares source expectations against landed database values. ## Core Features & Use Cases - Before/After Checklist: Verifies profiles, recipes, aliases, pools, customer tags, and saved-sheet reconciliation with PASS/WARN/FAIL status per area. - Safety and Evidence Rules: Enforces read-only dev-database queries, minimal stable identifiers, and redaction of credentials or personal data in reports. - Standard Audit Report: Produces a compact structured report with source evidence, landed values, confidence levels, and handoff recommendations. - Use Case: After importing a new pizza brand's workbook, run the audit to confirm dough weights match die sizes, recipe links resolve to existing pool rows, and no other customer's data was renamed or re-scoped. ## Quick Start Audit the most recent customer workbook import and report whether profiles, recipes, pools, and tags landed correctly.

Frequently Asked Questions about customer-import-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify a customer spec workbook import worked correctly?

Run a post-import audit comparing source expectations against landed database values for profiles, recipes, aliases, pools, and customer tags. Record PASS, WARN, or FAIL per area with the query used and a stable evidence identifier.

What should I check after importing a new customer workbook?

Check that profile weights are sane per die size, recipe names link to existing pool rows instead of duplicates, no cross-brand collisions occurred, applicator slots are ordered correctly, and saved-sheet reconciliation matches the applied parse.

Does the import audit modify production data?

No, the audit is strictly read-only and runs against the dev database via psql. It never edits production data, and a failed check triggers escalation or handoff rather than an automatic repair.

What happens when an import audit finds a mismatch?

A source-versus-landed mismatch is handed off to import-bug-investigation for layer classification before any fix is proposed. Incorrect persisted data goes to the data-heal-playbook with affected-row scope and preservation evidence.

Why might a landed value differ from the workbook source?

Linking, canonicalization, and blank-fill rules may intentionally transform values during import. A difference from the source is not automatically wrong; compare against both source evidence and the pre-import value before flagging.