tesserakit-schema

Catalog and lint JSON Schema documents.

tessera-schema discovers JSON Schema files, extracts their structure ($id, dialect, type, properties, required, $defs), and flags structural problems. It reads schemas only: no validation of instance data, no network.

Lint

tessera schema lint --input schemas/ --output ./out/schema_pack
tessera schema lint --input person.schema.json --output ./out/schema_pack

A JSON file is treated as a schema when it has $schema, properties, $defs/definitions, or a schema-style root type. Common non-schema files (package.json, tsconfig.json, ...) are skipped.

Artifacts written:

schemas.jsonl            one SchemaDoc per schema (type, properties, required, defs)
index.md                 the schema catalog
validation_report.md     structural findings
coverage_report.md       root-type distribution + title/$schema coverage

Findings

• required_not_in_properties (error) — a required field is not declared in properties
• missing_type (warning) — root schema has no type
• object_without_properties (warning) — type: object with no properties
• additional_properties_unset (info) — additionalProperties not set; the object is open by default
• missing_schema_version (info) — no $schema dialect
• missing_title (info)
• parse_error, no_schemas