CRM Data Hygiene Auditor

What it takes in

The inputs you provide on each run, validated before anything executes.

2 tasks, run in order.

Each task passes its result to the next. Here's what each does and what it produces.

1. 01
   Read all records from the CRM Contacts table and audit them for data-quality issues. Using today's date as the reference, detect: missing key fields (name, email, company, owner), malformed emails (no @, missing domain or top-level domain, or stray characters such as [at]), likely duplicates (the same person and company appearing more than once), stale records (Last Contacted far in the past, for example over a year ago), and placeholder or test junk (values like 'test', 'unknown', or '000'). If {{rules}} is provided, apply those thresholds or rules. Use only data in the table and identify records by name and company. Present readably, never JSON.

   Expected output

   A readable audit grouped by issue type, no JSON or code blocks: sections for Missing Fields, Malformed Emails, Likely Duplicates, Stale Records, and Placeholder or Junk, each listing the specific records (by name and company) and the exact problem. Include a one-line count of how many records are clean.
2. 02
   Turn the Task 1 audit into a prioritized cleanup plan. Order the issues by how much they hurt data reliability (for example duplicates and malformed emails before cosmetic gaps), and for each give the specific fix. Suggest a sensible sequence for working through them. Use only findings from Task 1. Present readably, never JSON.

   Expected output

   A readable cleanup plan, no JSON or code blocks: a prioritized list where each item names the record or issue, the recommended fix, and why it matters, followed by a short suggested order of operations.

What you can do with it