Bugfixes and documentation cleanup for v1.0 release.

internal/artifact/transcript.go

@@ -57,12 +57,12 @@ func FromMerged(cfg config.Config, merged model.MergedTranscript) schema.Transcr
 	}
 }
 
-// DefaultFromMerged converts the internal merged transcript model into the
-// compact default public serialized output contract.
-func DefaultFromMerged(cfg config.Config, merged model.MergedTranscript) schema.DefaultTranscript {
-	segments := make([]schema.DefaultSegment, len(merged.Segments))
+// IntermediateFromMerged converts the internal merged transcript model into the
+// compact intermediate public serialized output contract.
+func IntermediateFromMerged(cfg config.Config, merged model.MergedTranscript) schema.IntermediateTranscript {
+	segments := make([]schema.IntermediateSegment, len(merged.Segments))
 	for index, segment := range merged.Segments {
-		segments[index] = schema.DefaultSegment{
+		segments[index] = schema.IntermediateSegment{
 			ID:         segment.ID,
 			Start:      segment.Start,
 			End:        segment.End,
@@ -72,11 +72,11 @@ func DefaultFromMerged(cfg config.Config, merged model.MergedTranscript) schema.
 		}
 	}
 
-	return schema.DefaultTranscript{
-		Metadata: schema.DefaultMetadata{
+	return schema.IntermediateTranscript{
+		Metadata: schema.IntermediateMetadata{
 			Application:  ApplicationName,
 			Version:      buildinfo.Version,
-			OutputSchema: config.OutputSchemaDefault,
+			OutputSchema: config.OutputSchemaIntermediate,
 		},
 		Segments: segments,
 	}
@@ -110,14 +110,14 @@ func MinimalFromMerged(cfg config.Config, merged model.MergedTranscript) schema.
 // runtime-selected public output contract.
 func SelectedFromMerged(cfg config.Config, merged model.MergedTranscript) any {
 	switch cfg.OutputSchema {
-	case config.OutputSchemaDefault:
-		return DefaultFromMerged(cfg, merged)
 	case config.OutputSchemaMinimal:
 		return MinimalFromMerged(cfg, merged)
-	case config.OutputSchemaSeriatim:
+	case config.OutputSchemaFull:
 		return FromMerged(cfg, merged)
+	case config.OutputSchemaIntermediate:
+		return IntermediateFromMerged(cfg, merged)
 	default:
-		return FromMerged(cfg, merged)
+		return IntermediateFromMerged(cfg, merged)
 	}
 }
 

internal/artifact/transcript_test.go

@@ -23,28 +23,28 @@ func TestFromMergedUsesBuildVersion(t *testing.T) {
 	}
 }
 
-func TestSelectedFromMergedDefaultsToSeriatimTranscript(t *testing.T) {
+func TestSelectedFromMergedDefaultsToIntermediateTranscript(t *testing.T) {
 	got := SelectedFromMerged(config.Config{}, model.MergedTranscript{})
+	if _, ok := got.(schema.IntermediateTranscript); !ok {
+		t.Fatalf("selected artifact type = %T, want schema.IntermediateTranscript", got)
+	}
+}
+
+func TestSelectedFromMergedUsesIntermediateWhenConfigured(t *testing.T) {
+	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaIntermediate}, model.MergedTranscript{})
+	if _, ok := got.(schema.IntermediateTranscript); !ok {
+		t.Fatalf("selected artifact type = %T, want schema.IntermediateTranscript", got)
+	}
+}
+
+func TestSelectedFromMergedUsesFullWhenConfigured(t *testing.T) {
+	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaFull}, model.MergedTranscript{})
 	if _, ok := got.(schema.Transcript); !ok {
 		t.Fatalf("selected artifact type = %T, want schema.Transcript", got)
 	}
 }
 
-func TestSelectedFromMergedUsesDefaultWhenConfigured(t *testing.T) {
-	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaDefault}, model.MergedTranscript{})
-	if _, ok := got.(schema.DefaultTranscript); !ok {
-		t.Fatalf("selected artifact type = %T, want schema.DefaultTranscript", got)
-	}
-}
-
-func TestSelectedFromMergedUsesSeriatimWhenConfigured(t *testing.T) {
-	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaSeriatim}, model.MergedTranscript{})
-	if _, ok := got.(schema.Transcript); !ok {
-		t.Fatalf("selected artifact type = %T, want schema.Transcript", got)
-	}
-}
-
-func TestDefaultFromMergedEmitsOnlyDefaultShape(t *testing.T) {
+func TestIntermediateFromMergedEmitsOnlyIntermediateShape(t *testing.T) {
 	merged := model.MergedTranscript{
 		Segments: []model.Segment{
 			{
@@ -65,19 +65,19 @@ func TestDefaultFromMergedEmitsOnlyDefaultShape(t *testing.T) {
 		},
 	}
 
-	got := DefaultFromMerged(config.Config{OutputSchema: config.OutputSchemaDefault}, merged)
-	want := schema.DefaultTranscript{
-		Metadata: schema.DefaultMetadata{
+	got := IntermediateFromMerged(config.Config{OutputSchema: config.OutputSchemaIntermediate}, merged)
+	want := schema.IntermediateTranscript{
+		Metadata: schema.IntermediateMetadata{
 			Application:  ApplicationName,
 			Version:      buildinfo.Version,
-			OutputSchema: config.OutputSchemaDefault,
+			OutputSchema: config.OutputSchemaIntermediate,
 		},
-		Segments: []schema.DefaultSegment{
+		Segments: []schema.IntermediateSegment{
 			{ID: 1, Start: 1, End: 2, Speaker: "Alice", Text: "hello", Categories: []string{"backchannel"}},
 		},
 	}
 	if !reflect.DeepEqual(got, want) {
-		t.Fatalf("default transcript = %#v, want %#v", got, want)
+		t.Fatalf("intermediate transcript = %#v, want %#v", got, want)
 	}
 }
 

internal/builtin/postprocess.go

@@ -51,8 +51,8 @@ func (validateOutput) Process(ctx context.Context, in model.MergedTranscript, cf
 	selected := artifact.SelectedFromMerged(cfg, in)
 	var err error
 	switch transcript := selected.(type) {
-	case schema.DefaultTranscript:
-		err = schema.ValidateDefaultTranscript(transcript)
+	case schema.IntermediateTranscript:
+		err = schema.ValidateIntermediateTranscript(transcript)
 	case schema.MinimalTranscript:
 		err = schema.ValidateMinimalTranscript(transcript)
 	case schema.Transcript:

internal/builtin/postprocess_test.go

@@ -81,7 +81,7 @@ func TestValidateOutputUsesMinimalSchemaWhenConfigured(t *testing.T) {
 	}
 }
 
-func TestValidateOutputUsesSeriatimSchemaWhenConfigured(t *testing.T) {
+func TestValidateOutputUsesFullSchemaWhenConfigured(t *testing.T) {
 	merged := model.MergedTranscript{
 		Segments: []model.Segment{
 			{
@@ -100,7 +100,7 @@ func TestValidateOutputUsesSeriatimSchemaWhenConfigured(t *testing.T) {
 	}
 
 	cfg := testConfig()
-	cfg.OutputSchema = config.OutputSchemaSeriatim
+	cfg.OutputSchema = config.OutputSchemaFull
 	got, events, err := validateOutput{}.Process(context.Background(), merged, cfg)
 	if err != nil {
 		t.Fatalf("validate output: %v", err)

internal/cli/merge.go

@@ -15,7 +15,12 @@ func newMergeCommand() *cobra.Command {
 		Use:   "merge",
 		Short: "Run the transcript merge pipeline",
 		RunE: func(cmd *cobra.Command, args []string) error {
-			cfg, err := config.NewMergeConfig(opts)
+			mergeOpts := opts
+			if !cmd.Flags().Changed("output-schema") {
+				mergeOpts.OutputSchema = ""
+			}
+
+			cfg, err := config.NewMergeConfig(mergeOpts)
 			if err != nil {
 				return err
 			}
@@ -32,7 +37,7 @@ func newMergeCommand() *cobra.Command {
 	flags.StringVar(&opts.AutocorrectFile, "autocorrect", "", "autocorrect rules file")
 	flags.StringVar(&opts.InputReader, "input-reader", config.DefaultInputReader, "input reader module")
 	flags.StringVar(&opts.OutputModules, "output-modules", config.DefaultOutputModules, "comma-separated output modules")
-	flags.StringVar(&opts.OutputSchema, "output-schema", config.DefaultOutputSchema, "output JSON schema: default, minimal, or seriatim")
+	flags.StringVar(&opts.OutputSchema, "output-schema", config.DefaultOutputSchema, "output JSON schema: seriatim-minimal, seriatim-intermediate (default), or seriatim-full")
 	flags.StringVar(&opts.PreprocessingModules, "preprocessing-modules", config.DefaultPreprocessingModules, "comma-separated preprocessing modules")
 	flags.StringVar(&opts.PostprocessingModules, "postprocessing-modules", config.DefaultPostprocessingModules, "comma-separated postprocessing modules")
 	flags.StringVar(&opts.CoalesceGap, "coalesce-gap", config.DefaultCoalesceGapValue, "maximum same-speaker gap in seconds for coalesce")

internal/cli/merge_test.go

@@ -114,7 +114,8 @@ func TestMergeWritesMergedOutputAndReport(t *testing.T) {
 	}
 }
 
-func TestMergeWritesDefaultOutputSchema(t *testing.T) {
+func TestMergeWritesIntermediateOutputSchema(t *testing.T) {
+	t.Setenv(config.OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeJSONFile(t, dir, "input.json", `{
 		"segments": [
@@ -131,10 +132,10 @@ func TestMergeWritesDefaultOutputSchema(t *testing.T) {
 		t.Fatalf("merge failed: %v", err)
 	}
 
-	var transcript schema.DefaultTranscript
+	var transcript schema.IntermediateTranscript
 	readJSON(t, output, &transcript)
-	if transcript.Metadata.OutputSchema != config.OutputSchemaDefault {
-		t.Fatalf("output_schema = %q, want default", transcript.Metadata.OutputSchema)
+	if transcript.Metadata.OutputSchema != config.OutputSchemaIntermediate {
+		t.Fatalf("output_schema = %q, want %q", transcript.Metadata.OutputSchema, config.OutputSchemaIntermediate)
 	}
 	if len(transcript.Segments) != 1 {
 		t.Fatalf("segment count = %d, want 1", len(transcript.Segments))
@@ -149,12 +150,12 @@ func TestMergeWritesDefaultOutputSchema(t *testing.T) {
 	}
 	for _, forbidden := range []string{"overlap_groups", "source", "derived_from", "words"} {
 		if strings.Contains(string(outputBytes), forbidden) {
-			t.Fatalf("default output contains %q:\n%s", forbidden, outputBytes)
+			t.Fatalf("intermediate output contains %q:\n%s", forbidden, outputBytes)
 		}
 	}
 }
 
-func TestMergeWritesSeriatimOutputSchema(t *testing.T) {
+func TestMergeWritesFullOutputSchema(t *testing.T) {
 	dir := t.TempDir()
 	input := writeJSONFile(t, dir, "input.json", `{
 		"segments": [
@@ -166,7 +167,7 @@ func TestMergeWritesSeriatimOutputSchema(t *testing.T) {
 	err := executeMergeRaw(
 		"--input-file", input,
 		"--output-file", output,
-		"--output-schema", "seriatim",
+		"--output-schema", config.OutputSchemaFull,
 	)
 	if err != nil {
 		t.Fatalf("merge failed: %v", err)
@@ -203,7 +204,7 @@ func TestMergeWritesMinimalOutputSchema(t *testing.T) {
 	err := executeMerge(
 		"--input-file", input,
 		"--output-file", output,
-		"--output-schema", "minimal",
+		"--output-schema", config.OutputSchemaMinimal,
 		"--report-file", reportPath,
 	)
 	if err != nil {
@@ -215,8 +216,8 @@ func TestMergeWritesMinimalOutputSchema(t *testing.T) {
 	if transcript.Metadata.Application != "seriatim" {
 		t.Fatalf("application = %q, want seriatim", transcript.Metadata.Application)
 	}
-	if transcript.Metadata.OutputSchema != "minimal" {
-		t.Fatalf("output_schema = %q, want minimal", transcript.Metadata.OutputSchema)
+	if transcript.Metadata.OutputSchema != config.OutputSchemaMinimal {
+		t.Fatalf("output_schema = %q, want %q", transcript.Metadata.OutputSchema, config.OutputSchemaMinimal)
 	}
 	if got, want := len(transcript.Segments), 2; got != want {
 		t.Fatalf("segment count = %d, want %d", got, want)
@@ -330,7 +331,7 @@ func TestMergeValidateMinimalOutputBeforeAssignIDsFails(t *testing.T) {
 	err := executeMerge(
 		"--input-file", input,
 		"--output-file", output,
-		"--output-schema", "minimal",
+		"--output-schema", config.OutputSchemaMinimal,
 		"--postprocessing-modules", "validate-output,assign-ids",
 	)
 	if err == nil {
@@ -472,16 +473,16 @@ func TestMergeResolvesOverlapGroupsWithWordRuns(t *testing.T) {
 
 	var transcript model.FinalTranscript
 	readJSON(t, output, &transcript)
-	if len(transcript.OverlapGroups) != 0 {
-		t.Fatalf("overlap groups = %#v, want none", transcript.OverlapGroups)
+	if got, want := len(transcript.OverlapGroups), 1; got != want {
+		t.Fatalf("overlap group count = %d, want %d (%#v)", got, want, transcript.OverlapGroups)
 	}
-	if got, want := len(transcript.Segments), 3; got != want {
+	if got, want := len(transcript.Segments), 2; got != want {
 		t.Fatalf("segment count = %d, want %d", got, want)
 	}
 
-	wantTexts := []string{"outside hello there", "bob reply", "later"}
-	wantSpeakers := []string{"Alice", "Bob", "Alice"}
-	wantRefs := []string{"word-run:1:1:1", "word-run:1:2:1", "word-run:1:1:2"}
+	wantTexts := []string{"outside hello there later", "bob reply"}
+	wantSpeakers := []string{"Alice", "Bob"}
+	wantRefs := []string{"word-run:1:1:1", "word-run:1:2:1"}
 	for index, segment := range transcript.Segments {
 		if segment.ID != index+1 {
 			t.Fatalf("segment %d id = %d, want %d", index, segment.ID, index+1)
@@ -498,8 +499,8 @@ func TestMergeResolvesOverlapGroupsWithWordRuns(t *testing.T) {
 		if segment.SourceSegmentIndex != nil {
 			t.Fatalf("segment %d source_segment_index = %d, want nil", index, *segment.SourceSegmentIndex)
 		}
-		if segment.OverlapGroupID != 0 {
-			t.Fatalf("segment %d overlap_group_id = %d, want 0", index, segment.OverlapGroupID)
+		if segment.OverlapGroupID != 1 {
+			t.Fatalf("segment %d overlap_group_id = %d, want 1", index, segment.OverlapGroupID)
 		}
 	}
 	if !equalStrings(transcript.Segments[0].DerivedFrom, []string{inputA + "#0"}) {
@@ -516,7 +517,7 @@ func TestMergeResolvesOverlapGroupsWithWordRuns(t *testing.T) {
 
 	var rpt report.Report
 	readJSON(t, reportPath, &rpt)
-	if !hasReportEvent(rpt, "postprocessing", "resolve-overlaps", "processed 1 overlap group(s); changed 1; removed 2 original segment(s); created 3 replacement segment(s)") {
+	if !hasReportEvent(rpt, "postprocessing", "resolve-overlaps", "processed 1 overlap group(s); changed 1; removed 2 original segment(s); created 2 replacement segment(s)") {
 		t.Fatal("expected resolve-overlaps summary report event")
 	}
 }
@@ -1927,7 +1928,7 @@ func TestMergeResolveOverlapsPreservesAbsorbedContextPrefix(t *testing.T) {
 		"--input-file", inputA,
 		"--input-file", inputB,
 		"--speakers", speakers,
-		"--output-schema", "minimal",
+		"--output-schema", config.OutputSchemaMinimal,
 		"--output-file", output,
 	)
 	if err != nil {
@@ -1994,7 +1995,7 @@ func executeMerge(args ...string) error {
 	if !hasOutputSchemaFlag(args) {
 		// Most integration tests were written against the full envelope; keep
 		// that behavior unless the caller explicitly asks for another schema.
-		args = append(args, "--output-schema", config.OutputSchemaSeriatim)
+		args = append(args, "--output-schema", config.OutputSchemaFull)
 	}
 	return executeMergeRaw(args...)
 }

internal/config/config.go

@@ -13,22 +13,23 @@ import (
 const (
 	DefaultInputReader            = "json-files"
 	DefaultOutputModules          = "json"
-	DefaultOutputSchema           = OutputSchemaDefault
+	DefaultOutputSchema           = OutputSchemaIntermediate
 	DefaultPreprocessingModules   = "validate-raw,normalize-speakers,trim-text"
 	DefaultPostprocessingModules  = "detect-overlaps,resolve-overlaps,backchannel,filler,resolve-danglers,coalesce,detect-overlaps,autocorrect,assign-ids,validate-output"
-	DefaultOverlapWordRunGap      = 0.75
+	DefaultOverlapWordRunGap      = 1.0
 	DefaultWordRunReorderWindow   = 1.0
 	DefaultCoalesceGap            = 3.0
 	DefaultCoalesceGapValue       = "3.0"
 	DefaultBackchannelMaxDuration = 2.0
 	DefaultFillerMaxDuration      = 1.25
+	OutputSchemaEnv               = "SERIATIM_OUTPUT_SCHEMA"
 	OverlapWordRunGapEnv          = "SERIATIM_OVERLAP_WORD_RUN_GAP"
 	WordRunReorderWindowEnv       = "SERIATIM_OVERLAP_WORD_RUN_REORDER_WINDOW"
 	BackchannelMaxDurationEnv     = "SERIATIM_BACKCHANNEL_MAX_DURATION"
 	FillerMaxDurationEnv          = "SERIATIM_FILLER_MAX_DURATION"
-	OutputSchemaDefault           = "default"
-	OutputSchemaSeriatim          = "seriatim"
-	OutputSchemaMinimal           = "minimal"
+	OutputSchemaMinimal           = "seriatim-minimal"
+	OutputSchemaIntermediate      = "seriatim-intermediate"
+	OutputSchemaFull              = "seriatim-full"
 )
 
 // MergeOptions captures raw CLI option values before validation.
@@ -70,7 +71,7 @@ func NewMergeConfig(opts MergeOptions) (Config, error) {
 	cfg := Config{
 		InputReader:            strings.TrimSpace(opts.InputReader),
 		OutputModules:          nil,
-		OutputSchema:           strings.TrimSpace(opts.OutputSchema),
+		OutputSchema:           "",
 		PreprocessingModules:   nil,
 		PostprocessingModules:  nil,
 		OverlapWordRunGap:      DefaultOverlapWordRunGap,
@@ -83,14 +84,12 @@ func NewMergeConfig(opts MergeOptions) (Config, error) {
 	if cfg.InputReader == "" {
 		return Config{}, errors.New("--input-reader is required")
 	}
-	if cfg.OutputSchema == "" {
-		cfg.OutputSchema = DefaultOutputSchema
-	}
-	if err := validateOutputSchema(cfg.OutputSchema); err != nil {
+	var err error
+	cfg.OutputSchema, err = resolveOutputSchema(opts.OutputSchema)
+	if err != nil {
 		return Config{}, err
 	}
 
-	var err error
 	cfg.OutputModules, err = parseModuleList(opts.OutputModules)
 	if err != nil {
 		return Config{}, fmt.Errorf("--output-modules: %w", err)
@@ -189,13 +188,27 @@ func parseModuleList(value string) ([]string, error) {
 
 func validateOutputSchema(value string) error {
 	switch value {
-	case OutputSchemaDefault, OutputSchemaSeriatim, OutputSchemaMinimal:
+	case OutputSchemaMinimal, OutputSchemaIntermediate, OutputSchemaFull:
 		return nil
 	default:
-		return fmt.Errorf("--output-schema must be one of %q, %q, or %q", OutputSchemaDefault, OutputSchemaMinimal, OutputSchemaSeriatim)
+		return fmt.Errorf("--output-schema must be one of %q, %q, or %q", OutputSchemaMinimal, OutputSchemaIntermediate, OutputSchemaFull)
 	}
 }
 
+func resolveOutputSchema(value string) (string, error) {
+	value = strings.TrimSpace(value)
+	if value == "" {
+		value = strings.TrimSpace(os.Getenv(OutputSchemaEnv))
+	}
+	if value == "" {
+		value = DefaultOutputSchema
+	}
+	if err := validateOutputSchema(value); err != nil {
+		return "", err
+	}
+	return value, nil
+}
+
 func normalizeInputFiles(paths []string) ([]string, error) {
 	if len(paths) == 0 {
 		return nil, errors.New("at least one --input-file is required")

internal/config/config_test.go

@@ -46,7 +46,8 @@ func TestDuplicateInputFilesFailValidation(t *testing.T) {
 	}
 }
 
-func TestOutputSchemaDefaultsToDefault(t *testing.T) {
+func TestOutputSchemaDefaultsToIntermediate(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -67,7 +68,8 @@ func TestOutputSchemaDefaultsToDefault(t *testing.T) {
 	}
 }
 
-func TestOutputSchemaAcceptsDefault(t *testing.T) {
+func TestOutputSchemaAcceptsIntermediate(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -77,19 +79,20 @@ func TestOutputSchemaAcceptsDefault(t *testing.T) {
 		OutputFile:            output,
 		InputReader:           DefaultInputReader,
 		OutputModules:         DefaultOutputModules,
-		OutputSchema:          OutputSchemaDefault,
+		OutputSchema:          OutputSchemaIntermediate,
 		PreprocessingModules:  DefaultPreprocessingModules,
 		PostprocessingModules: DefaultPostprocessingModules,
 	})
 	if err != nil {
 		t.Fatalf("config failed: %v", err)
 	}
-	if cfg.OutputSchema != OutputSchemaDefault {
-		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaDefault)
+	if cfg.OutputSchema != OutputSchemaIntermediate {
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaIntermediate)
 	}
 }
 
 func TestOutputSchemaAcceptsMinimal(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -111,7 +114,8 @@ func TestOutputSchemaAcceptsMinimal(t *testing.T) {
 	}
 }
 
-func TestOutputSchemaAcceptsSeriatim(t *testing.T) {
+func TestOutputSchemaAcceptsFull(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -121,19 +125,87 @@ func TestOutputSchemaAcceptsSeriatim(t *testing.T) {
 		OutputFile:            output,
 		InputReader:           DefaultInputReader,
 		OutputModules:         DefaultOutputModules,
-		OutputSchema:          OutputSchemaSeriatim,
+		OutputSchema:          OutputSchemaFull,
 		PreprocessingModules:  DefaultPreprocessingModules,
 		PostprocessingModules: DefaultPostprocessingModules,
 	})
 	if err != nil {
 		t.Fatalf("config failed: %v", err)
 	}
-	if cfg.OutputSchema != OutputSchemaSeriatim {
-		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaSeriatim)
+	if cfg.OutputSchema != OutputSchemaFull {
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaFull)
+	}
+}
+
+func TestOutputSchemaUsesEnvWhenFlagOmitted(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, OutputSchemaFull)
+	dir := t.TempDir()
+	input := writeTempFile(t, dir, "input.json")
+	output := filepath.Join(dir, "merged.json")
+
+	cfg, err := NewMergeConfig(MergeOptions{
+		InputFiles:            []string{input},
+		OutputFile:            output,
+		InputReader:           DefaultInputReader,
+		OutputModules:         DefaultOutputModules,
+		PreprocessingModules:  DefaultPreprocessingModules,
+		PostprocessingModules: DefaultPostprocessingModules,
+	})
+	if err != nil {
+		t.Fatalf("config failed: %v", err)
+	}
+	if cfg.OutputSchema != OutputSchemaFull {
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaFull)
+	}
+}
+
+func TestOutputSchemaFlagOverridesEnv(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, OutputSchemaFull)
+	dir := t.TempDir()
+	input := writeTempFile(t, dir, "input.json")
+	output := filepath.Join(dir, "merged.json")
+
+	cfg, err := NewMergeConfig(MergeOptions{
+		InputFiles:            []string{input},
+		OutputFile:            output,
+		InputReader:           DefaultInputReader,
+		OutputModules:         DefaultOutputModules,
+		OutputSchema:          OutputSchemaMinimal,
+		PreprocessingModules:  DefaultPreprocessingModules,
+		PostprocessingModules: DefaultPostprocessingModules,
+	})
+	if err != nil {
+		t.Fatalf("config failed: %v", err)
+	}
+	if cfg.OutputSchema != OutputSchemaMinimal {
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaMinimal)
+	}
+}
+
+func TestOutputSchemaRejectsInvalidEnvValue(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "compact")
+	dir := t.TempDir()
+	input := writeTempFile(t, dir, "input.json")
+	output := filepath.Join(dir, "merged.json")
+
+	_, err := NewMergeConfig(MergeOptions{
+		InputFiles:            []string{input},
+		OutputFile:            output,
+		InputReader:           DefaultInputReader,
+		OutputModules:         DefaultOutputModules,
+		PreprocessingModules:  DefaultPreprocessingModules,
+		PostprocessingModules: DefaultPostprocessingModules,
+	})
+	if err == nil {
+		t.Fatal("expected output schema error")
+	}
+	if !strings.Contains(err.Error(), "--output-schema must be one of") {
+		t.Fatalf("unexpected error: %v", err)
 	}
 }
 
 func TestOutputSchemaRejectsUnknownValue(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -155,7 +227,36 @@ func TestOutputSchemaRejectsUnknownValue(t *testing.T) {
 	}
 }
 
-func TestOverlapWordRunGapDefaultsTo075(t *testing.T) {
+func TestOutputSchemaRejectsLegacyValues(t *testing.T) {
+	tests := []string{"default", "minimal", "seriatim"}
+
+	for _, legacy := range tests {
+		t.Run(legacy, func(t *testing.T) {
+			t.Setenv(OutputSchemaEnv, "")
+			dir := t.TempDir()
+			input := writeTempFile(t, dir, "input.json")
+			output := filepath.Join(dir, "merged.json")
+
+			_, err := NewMergeConfig(MergeOptions{
+				InputFiles:            []string{input},
+				OutputFile:            output,
+				InputReader:           DefaultInputReader,
+				OutputModules:         DefaultOutputModules,
+				OutputSchema:          legacy,
+				PreprocessingModules:  DefaultPreprocessingModules,
+				PostprocessingModules: DefaultPostprocessingModules,
+			})
+			if err == nil {
+				t.Fatal("expected output schema error")
+			}
+			if !strings.Contains(err.Error(), "--output-schema must be one of") {
+				t.Fatalf("unexpected error: %v", err)
+			}
+		})
+	}
+}
+
+func TestOverlapWordRunGapDefaultsTo1(t *testing.T) {
 	t.Setenv(OverlapWordRunGapEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")

internal/danglers/danglers.go

@@ -184,7 +184,7 @@ func nearestPriorMatch(segments []model.Segment, consumed []bool, index int) int
 		if consumed[candidate] {
 			continue
 		}
-		if sharesDerivedFrom(segments[index], segments[candidate]) {
+		if segments[index].Speaker == segments[candidate].Speaker && sharesDerivedFrom(segments[index], segments[candidate]) {
 			return candidate
 		}
 	}
@@ -196,7 +196,7 @@ func nearestSubsequentMatch(segments []model.Segment, consumed []bool, index int
 		if consumed[candidate] {
 			continue
 		}
-		if sharesDerivedFrom(segments[index], segments[candidate]) {
+		if segments[index].Speaker == segments[candidate].Speaker && sharesDerivedFrom(segments[index], segments[candidate]) {
 			return candidate
 		}
 	}

internal/danglers/danglers_test.go

@@ -51,6 +51,30 @@ func TestApplyUsesAnyDerivedFromIntersection(t *testing.T) {
 	assertSegment(t, got.Segments[0], "resolve-danglers:1", "target end.", 1, 4, []string{"source#1", "source#2", "source#3"})
 }
 
+func TestApplyDoesNotMergeDanglersAcrossSpeakersBackward(t *testing.T) {
+	in := transcript(
+		segment("a", "Alice", 1, 2, "target", []string{"source#1"}),
+		segment("b", "Bob", 3, 4, "end.", []string{"source#1"}),
+	)
+
+	got, summary := Apply(in)
+	if summary.DanglersMerged != 0 || !reflect.DeepEqual(got, in) {
+		t.Fatalf("unexpected merge:\ngot  %#v\nwant %#v", got, in)
+	}
+}
+
+func TestApplyDoesNotMergeDanglersAcrossSpeakersForward(t *testing.T) {
+	in := transcript(
+		segment("a", "Alice", 1, 2, "start", []string{"source#1"}),
+		segment("b", "Bob", 3, 4, "target", []string{"source#1"}),
+	)
+
+	got, summary := Apply(in)
+	if summary.DanglersMerged != 0 || !reflect.DeepEqual(got, in) {
+		t.Fatalf("unexpected merge:\ngot  %#v\nwant %#v", got, in)
+	}
+}
+
 func TestApplyDoesNotMergeWithoutSharedProvenance(t *testing.T) {
 	in := transcript(
 		segment("a", "Alice", 1, 2, "target", []string{"source#1"}),