lint-consistency

To identify duplicate cohorts in lint_scan.json, evaluate candidate groups sharing the same parent_institution and modality to classify them as true duplicates, distinct waves, or separate entities, generating a non-destructive decision report.

Resolving lint consistency candidates safely involves a read-only process that scans consistency sections and entity bodies to produce a merge decision report at a specified out_path, ensuring source data remains completely unmodified.

Lint consistency classification distinguishes duplicates from distinct waves by evaluating candidate cohorts within lint_scan.json that share matching parent_institution and modality fields, then applying rules to determine if they are true duplicates or separate platforms.

Yes, you can generate a merge decision report without altering wiki data because the process operates strictly in read-only mode, reading the consistency section and entity bodies to write a single report to out_path.

A key limitation is that lint consistency evaluation only groups candidate cohorts sharing the exact same parent_institution and modality, meaning cohorts with differing modalities or parents are not classified as potential merge candidates.

You need a lint_scan.json file containing a consistency section and entity bodies with parent_institution and modality fields, as the read-only process reads these specific structures to classify cohorts and output a merge recommendation report.