Design an agent that classifies uploaded documents and suggests traceable revisions, keeping deterministic rules separate from model judgment and letting a human override.

Split the work by what can be checked. A classifier labels the document type, and the type selects a rule set. Checkable rules run as code: required clauses present, dates valid, amounts consistent. The model handles judgment rules, and every finding cites the rule id, the exact span and a proposed edit. Suggestions arrive as a reviewable diff; humans accept, reject or override, and their decisions are logged as eval data.