Data Quality Exception Review with Claude Code: API Runtime Pattern
A production playbook for data quality exception review in cross-industry operations using Claude Code: api runtime pattern, run-scoped inputs, logs, typed results, and artifacts.
Audience: Data operations teams
The problem
Data operations teams need data quality exception review to run repeatedly against failed rows, schemas, validation rules, and samples. In cross-industry operations, the pain is not one good answer; it is repeatability, auditability, exception handling, and evidence that survives handoff.
Implementation path
Package the data quality exception review instructions as a skill, send failed rows, schemas, validation rules, and samples as run-scoped inputs, execute with Claude Code, poll terminal status, and consume argo.result.v1 instead of parsing a transcript.
Tradeoffs and failure modes
The API boundary forces the workflow to define inputs, terminal states, and result shape before customers depend on it. For data quality exception review, the practical test is whether a second run can be debugged, retried, and consumed by a product without reading the raw agent transcript.
Run request
POST /api/skills/<skill_id>/run
provider=claude-code
workflow=data-quality-exceptions
inputs[]=@./input-pack.zip
result_schema=argo.result.v1
Run this on Argo