Refactor validators into built-in registry chains

README.md

@@ -150,6 +150,7 @@ audita config print-effective --config audita.yml
 
 For full config-file schema and examples, see [`docs/configuration.md`](docs/configuration.md).
 For output-schema details, see [`docs/output-schemas.md`](docs/output-schemas.md).
+For built-in validator keys and chain definitions, see [`docs/validators.md`](docs/validators.md).
 For CLI/process compatibility guarantees, see [`docs/public-contract.md`](docs/public-contract.md).
 
 ### Modules

docs/architecture.md

@@ -132,6 +132,10 @@ internal/framework/validators/
   llm_batching.go
   llm_validators.go
 
+internal/validators/
+  registry.go
+  chains.go
+
 internal/framework/llm/
   openai_compatible_client.go
   scheduler.go
@@ -428,6 +432,63 @@ These primitives are wired into the production runner and report model. The gram
 
 Validator rejections are reported distinctly from proposal-application skips.
 
+Validator composition is now explicit and registry-backed through `internal/validators`:
+- built-in validator registry with stable keys and lookup/build failure for unknown keys;
+- built-in chain definitions per production module key;
+- production modules resolve validator chains from those built-in definitions.
+
+Stable built-in validator keys:
+- deterministic:
+  - `confidence_threshold`
+  - `original_text_presence`
+  - `non_empty_corrected_text`
+  - `no_effect`
+  - `protected_terms`
+- LLM-backed:
+  - `spoken_form_plausibility`
+  - `meaning_reversal_review`
+  - `editorial_review`
+  - `grammar_review`
+  - `spoken_word_review`
+
+Built-in module chains:
+- `glossary`:
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `spoken_form_plausibility`
+  - `meaning_reversal_review`
+- `homophones`:
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `spoken_form_plausibility`
+  - `meaning_reversal_review`
+- `spoken_word`:
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `spoken_word_review`
+  - `meaning_reversal_review`
+- `grammar`:
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `grammar_review`
+  - `meaning_reversal_review`
+
+1.0 boundary:
+- validator chains are built-in and not user-configurable from config/CLI.
+- existing threshold and batching knobs remain configurable.
+
 ## Implemented LLM-backed validator infrastructure
 `internal/framework/validators` now includes LLM-backed validator support:
 - typed request/response models for structured LLM validation;
@@ -563,6 +624,7 @@ Current process reports also include:
 - run-level module summary totals and failed module instance metadata.
 - module-level validator decisions and validator rejections.
 - optional decision-level diagnostic artifact paths for validator LLM interactions when available.
+- stable validator keys in `validator_name` fields for validator decisions/rejections.
 - explicit report metadata:
   - report schema name;
   - report schema version;

docs/configuration.md

@@ -182,3 +182,4 @@ Current guidance:
 - prefer file config for baseline behavior;
 - keep environment variables for secrets/deployment-specific overrides;
 - use CLI flags for per-run overrides.
+- validator chains are built-in and are not user-configurable in config.

docs/public-contract.md

@@ -10,6 +10,7 @@ This contract covers:
 - transcript/glossary input forms
 - transcript output schema selection
 - process report schema metadata
+- stable validator key identifiers in report/diagnostics records
 - diagnostics directory behavior
 - stdout/stderr and exit-code behavior
 - secret redaction guarantees
@@ -89,6 +90,8 @@ Current values:
 
 `--report-json` output and diagnostics run-dir `report.json` use the same report schema metadata.
 
+Validator decision/rejection records in reports use stable validator keys in `validator_name`.
+
 ## Diagnostics directory behavior
 
 When diagnostics directory creation succeeds, Audita writes run artifacts including:

docs/roadmap.md

@@ -511,6 +511,20 @@ Add explicit report schema metadata, for example:
 
 Make validators as easy to reason about as modules.
 
+## Implementation status (2026-05-13)
+
+This workstream is now implemented for built-in validator composition:
+- first-class built-in validator registry exists in `internal/validators`;
+- stable validator keys are defined and used for built-in chains and runtime validator names;
+- explicit built-in module validator-chain definitions are implemented and resolved through registry-backed chain wiring;
+- production module constructors use built-in chain resolution instead of ad hoc manual validator construction;
+- runner execution preserves deterministic ordering (deterministic validators before LLM-backed validators);
+- report validator decision/rejection records now carry stable validator keys.
+
+Current boundary:
+- validator chains are built-in and not user-configurable.
+- prompt-asset registries, prompt metadata, scheduler utilization diagnostics, correction ledgers, and generated summaries remain planned.
+
 Validators are now central runtime components. They are reused across modules, have deterministic and LLM-backed implementations, produce diagnostics, and affect final correction acceptance. They should therefore have stable identities, registry metadata, and composable chain definitions.
 
 ## Package structure

docs/validators.md

@@ -0,0 +1,111 @@
+# Audita Validators
+
+This document describes Audita's built-in validator registry and module validator chains.
+
+## Scope
+
+Validator chains are built-in runtime behavior.
+
+Current 1.0 boundary:
+- built-in validator keys and built-in module chains are stable runtime identifiers;
+- thresholds and batching knobs remain configurable where already supported;
+- arbitrary user-defined validator chains are deferred.
+
+## Built-in validator keys
+
+### Deterministic validators
+
+- `confidence_threshold`
+  - checks proposal confidence against module-specific configured threshold.
+- `original_text_presence`
+  - ensures target segment exists and `original_text` exists in current working segment text.
+- `non_empty_corrected_text`
+  - rejects blank/whitespace-only `corrected_text`.
+- `no_effect`
+  - rejects proposals where `original_text == corrected_text`.
+- `protected_terms`
+  - protects glossary-derived terms from unsafe mutations in non-glossary modules.
+  - glossary stages use glossary-specific protection logic but still report this same stable key.
+
+### LLM-backed validators
+
+- `spoken_form_plausibility`
+  - checks whether proposed spoken-form change remains plausible in transcript context.
+- `meaning_reversal_review`
+  - checks for likely meaning reversal or semantic contradiction.
+- `editorial_review`
+  - performs conservative editorial safety review.
+- `grammar_review`
+  - checks grammar-stage proposals for grammar-focused safety constraints.
+- `spoken_word_review`
+  - checks spoken-word-stage proposals for dysfluency-cleanup safety constraints.
+
+## Built-in module chains
+
+Current built-in chains resolved from `internal/validators/chains.go`:
+
+- `glossary`
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `spoken_form_plausibility`
+  - `meaning_reversal_review`
+
+- `homophones`
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `spoken_form_plausibility`
+  - `meaning_reversal_review`
+
+- `spoken_word`
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `spoken_word_review`
+  - `meaning_reversal_review`
+
+- `grammar`
+  - `no_effect`
+  - `original_text_presence`
+  - `confidence_threshold`
+  - `protected_terms`
+  - `non_empty_corrected_text`
+  - `grammar_review`
+  - `meaning_reversal_review`
+
+## Execution semantics
+
+- modules execute serially;
+- section proposal work can run concurrently within a module;
+- deterministic validators run before LLM-backed validators;
+- malformed/missing/duplicate/unknown LLM validator decisions fail safely;
+- approved proposals are applied once per module after section work settles.
+
+## Validator rejections vs proposal-application skips
+
+- validator rejection:
+  - proposal is denied by validator-chain review and appears in validator rejection reporting with validator key and reason code.
+- proposal-application skip:
+  - proposal passed validators but could not be applied under replacement-policy semantics (for example no matching span at apply time).
+
+These are separate outcomes and are reported separately.
+
+## Reporting and diagnostics identity
+
+- report validator decision/rejection entries use stable validator keys in `validator_name`.
+- validator LLM diagnostics include validator identity in interaction metadata and structured response schema metadata.
+
+## Configurable knobs that remain supported
+
+- per-module confidence thresholds (`thresholds.*` / equivalent env+CLI overrides)
+- validation batching limits (`validation_max_prompt_tokens` / equivalent env+CLI overrides)
+- validation LLM model/base URL/timeout/retries/concurrency settings
+
+These tune validator behavior without exposing arbitrary user-defined chains.

internal/cli/run_test.go

@@ -3048,6 +3048,9 @@ func TestRunProcessExplicitHomophonesProtectedGlossaryTermRejected(t *testing.T)
 	if module.ValidatorRejected[0].ReasonCode != validators.ReasonProtectedGlossaryTerm {
 		t.Fatalf("expected protected glossary term rejection, got %+v", module.ValidatorRejected[0])
 	}
+	if module.ValidatorRejected[0].ValidatorName != "protected_terms" {
+		t.Fatalf("expected stable validator key protected_terms, got %+v", module.ValidatorRejected[0])
+	}
 }
 
 func TestRunProcessExplicitHomophonesMalformedLLMOutputFailsWithErrorLog(t *testing.T) {

internal/cli/testdata/parity/llm-validator-rejection.case.json

@@ -14,6 +14,6 @@
     "total_skipped_changes": 1,
     "validator_rejected_reason_codes": ["llm_rejected"],
     "expected_proposal_calls": ["grammar:proposal"],
-    "expected_validation_calls": ["grammar:section-0000:grammar_only_guard:batch-0000"]
+    "expected_validation_calls": ["grammar:section-0000:grammar_review:batch-0000"]
   }
 }

internal/framework/validators/deterministic.go

@@ -59,7 +59,7 @@ func (v OriginalTextPresenceValidator) Validate(_ context.Context, req Request)
 
 type NonEmptyCorrectionValidator struct{}
 
-func (v NonEmptyCorrectionValidator) Name() string { return "non_empty_correction" }
+func (v NonEmptyCorrectionValidator) Name() string { return "non_empty_corrected_text" }
 
 func (v NonEmptyCorrectionValidator) Validate(_ context.Context, req Request) (Result, error) {
 	decisions := make([]Decision, 0, len(req.CandidateProposal))
@@ -97,7 +97,7 @@ func (v NoEffectValidator) Validate(_ context.Context, req Request) (Result, err
 
 type ProtectedGlossaryTermValidator struct{}
 
-func (v ProtectedGlossaryTermValidator) Name() string { return "protected_glossary_terms" }
+func (v ProtectedGlossaryTermValidator) Name() string { return "protected_terms" }
 
 func (v ProtectedGlossaryTermValidator) Validate(_ context.Context, req Request) (Result, error) {
 	if req.ModuleKey == "glossary" {
@@ -127,7 +127,7 @@ func (v ProtectedGlossaryTermValidator) Validate(_ context.Context, req Request)
 type GlossaryStageProtectedGlossaryTermValidator struct{}
 
 func (v GlossaryStageProtectedGlossaryTermValidator) Name() string {
-	return "glossary_stage_protected_glossary_terms"
+	return "protected_terms"
 }
 
 func (v GlossaryStageProtectedGlossaryTermValidator) Validate(_ context.Context, req Request) (Result, error) {

internal/modules/glossary/module.go

@@ -8,7 +8,7 @@ import (
 	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposal_generation"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposals"
-	"gitea.maximumdirect.net/eric/audita/internal/framework/validators"
+	builtinvalidators "gitea.maximumdirect.net/eric/audita/internal/validators"
 )
 
 type Module struct {
@@ -16,25 +16,12 @@ type Module struct {
 }
 
 func New() (*Module, error) {
-	spokenForm, err := validators.NewLLMBackedValidator("spoken_form_plausibility_review", validators.LLMValidatorTypeSpokenFormPlausibility, "")
+	chain, err := builtinvalidators.ResolveBuiltInChain("glossary", builtinvalidators.NewBuiltInRegistry())
 	if err != nil {
 		return nil, err
 	}
-	meaningReversal, err := validators.NewLLMBackedValidator("meaning_reversal_review", validators.LLMValidatorTypeMeaningReversal, "")
-	if err != nil {
-		return nil, err
-	}
-
 	return &Module{
-		validators: []contracts.Validator{
+		validators: chain,
-			validators.NoEffectValidator{},
-			validators.OriginalTextPresenceValidator{},
-			validators.ConfidenceThresholdValidator{},
-			validators.GlossaryStageProtectedGlossaryTermValidator{},
-			validators.NonEmptyCorrectionValidator{},
-			spokenForm,
-			meaningReversal,
-		},
 	}, nil
 }
 

internal/modules/glossary/module_test.go

@@ -132,9 +132,9 @@ func TestGlossaryModuleValidatorChain(t *testing.T) {
 		"no_effect",
 		"original_text_presence",
 		"confidence_threshold",
-		"glossary_stage_protected_glossary_terms",
+		"protected_terms",
-		"non_empty_correction",
+		"non_empty_corrected_text",
-		"spoken_form_plausibility_review",
+		"spoken_form_plausibility",
 		"meaning_reversal_review",
 	}
 	if strings.Join(got, ",") != strings.Join(want, ",") {

internal/modules/grammar/module.go

@@ -8,7 +8,7 @@ import (
 	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposal_generation"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposals"
-	"gitea.maximumdirect.net/eric/audita/internal/framework/validators"
+	builtinvalidators "gitea.maximumdirect.net/eric/audita/internal/validators"
 )
 
 type Module struct {
@@ -16,25 +16,12 @@ type Module struct {
 }
 
 func New() (*Module, error) {
-	grammarOnlyGuard, err := validators.NewLLMBackedValidator("grammar_only_guard", validators.LLMValidatorTypeGrammarReview, "")
+	chain, err := builtinvalidators.ResolveBuiltInChain("grammar", builtinvalidators.NewBuiltInRegistry())
 	if err != nil {
 		return nil, err
 	}
-	meaningReversal, err := validators.NewLLMBackedValidator("meaning_reversal_review", validators.LLMValidatorTypeMeaningReversal, "")
-	if err != nil {
-		return nil, err
-	}
-
 	return &Module{
-		validators: []contracts.Validator{
+		validators: chain,
-			validators.NoEffectValidator{},
-			validators.OriginalTextPresenceValidator{},
-			validators.ConfidenceThresholdValidator{},
-			validators.ProtectedGlossaryTermValidator{},
-			validators.NonEmptyCorrectionValidator{},
-			grammarOnlyGuard,
-			meaningReversal,
-		},
 	}, nil
 }
 

internal/modules/grammar/module_test.go

@@ -134,9 +134,9 @@ func TestGrammarModuleValidatorChain(t *testing.T) {
 		"no_effect",
 		"original_text_presence",
 		"confidence_threshold",
-		"protected_glossary_terms",
+		"protected_terms",
-		"non_empty_correction",
+		"non_empty_corrected_text",
-		"grammar_only_guard",
+		"grammar_review",
 		"meaning_reversal_review",
 	}
 	if strings.Join(got, ",") != strings.Join(want, ",") {

internal/modules/homophones/module.go

@@ -8,7 +8,7 @@ import (
 	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposal_generation"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposals"
-	"gitea.maximumdirect.net/eric/audita/internal/framework/validators"
+	builtinvalidators "gitea.maximumdirect.net/eric/audita/internal/validators"
 )
 
 type Module struct {
@@ -16,25 +16,12 @@ type Module struct {
 }
 
 func New() (*Module, error) {
-	spokenForm, err := validators.NewLLMBackedValidator("spoken_form_plausibility_review", validators.LLMValidatorTypeSpokenFormPlausibility, "")
+	chain, err := builtinvalidators.ResolveBuiltInChain("homophones", builtinvalidators.NewBuiltInRegistry())
 	if err != nil {
 		return nil, err
 	}
-	meaningReversal, err := validators.NewLLMBackedValidator("meaning_reversal_review", validators.LLMValidatorTypeMeaningReversal, "")
-	if err != nil {
-		return nil, err
-	}
-
 	return &Module{
-		validators: []contracts.Validator{
+		validators: chain,
-			validators.NoEffectValidator{},
-			validators.OriginalTextPresenceValidator{},
-			validators.ConfidenceThresholdValidator{},
-			validators.ProtectedGlossaryTermValidator{},
-			validators.NonEmptyCorrectionValidator{},
-			spokenForm,
-			meaningReversal,
-		},
 	}, nil
 }
 

internal/modules/homophones/module_test.go

@@ -148,9 +148,9 @@ func TestHomophonesModuleValidatorChain(t *testing.T) {
 		"no_effect",
 		"original_text_presence",
 		"confidence_threshold",
-		"protected_glossary_terms",
+		"protected_terms",
-		"non_empty_correction",
+		"non_empty_corrected_text",
-		"spoken_form_plausibility_review",
+		"spoken_form_plausibility",
 		"meaning_reversal_review",
 	}
 	if strings.Join(got, ",") != strings.Join(want, ",") {
@@ -276,13 +276,13 @@ func TestHomophonesProtectedGlossaryTermBehaviorRejectsUnsafeCorrection(t *testi
 	}
 	var protectedValidator contracts.Validator
 	for _, v := range module.Validators() {
-		if v.Name() == "protected_glossary_terms" {
+		if v.Name() == "protected_terms" {
 			protectedValidator = v
 			break
 		}
 	}
 	if protectedValidator == nil {
-		t.Fatal("expected protected_glossary_terms validator")
+		t.Fatal("expected protected_terms validator")
 	}
 
 	working := &schema.Transcript{Segments: []schema.Segment{

internal/modules/spoken_word/module.go

@@ -8,7 +8,7 @@ import (
 	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposal_generation"
 	"gitea.maximumdirect.net/eric/audita/internal/framework/proposals"
-	"gitea.maximumdirect.net/eric/audita/internal/framework/validators"
+	builtinvalidators "gitea.maximumdirect.net/eric/audita/internal/validators"
 )
 
 type Module struct {
@@ -16,25 +16,12 @@ type Module struct {
 }
 
 func New() (*Module, error) {
-	spokenWordReview, err := validators.NewLLMBackedValidator("spoken_word_review", validators.LLMValidatorTypeSpokenWordReview, "")
+	chain, err := builtinvalidators.ResolveBuiltInChain("spoken_word", builtinvalidators.NewBuiltInRegistry())
 	if err != nil {
 		return nil, err
 	}
-	meaningReversal, err := validators.NewLLMBackedValidator("meaning_reversal_review", validators.LLMValidatorTypeMeaningReversal, "")
-	if err != nil {
-		return nil, err
-	}
-
 	return &Module{
-		validators: []contracts.Validator{
+		validators: chain,
-			validators.NoEffectValidator{},
-			validators.OriginalTextPresenceValidator{},
-			validators.ConfidenceThresholdValidator{},
-			validators.ProtectedGlossaryTermValidator{},
-			validators.NonEmptyCorrectionValidator{},
-			spokenWordReview,
-			meaningReversal,
-		},
 	}, nil
 }
 

internal/modules/spoken_word/module_test.go

@@ -136,8 +136,8 @@ func TestSpokenWordModuleValidatorChain(t *testing.T) {
 		"no_effect",
 		"original_text_presence",
 		"confidence_threshold",
-		"protected_glossary_terms",
+		"protected_terms",
-		"non_empty_correction",
+		"non_empty_corrected_text",
 		"spoken_word_review",
 		"meaning_reversal_review",
 	}
@@ -264,13 +264,13 @@ func TestSpokenWordProtectedGlossaryTermBehaviorRejectsUnsafeCorrection(t *testi
 	}
 	var protectedValidator contracts.Validator
 	for _, v := range module.Validators() {
-		if v.Name() == "protected_glossary_terms" {
+		if v.Name() == "protected_terms" {
 			protectedValidator = v
 			break
 		}
 	}
 	if protectedValidator == nil {
-		t.Fatal("expected protected_glossary_terms validator")
+		t.Fatal("expected protected_terms validator")
 	}
 
 	working := &schema.Transcript{Segments: []schema.Segment{

internal/validators/chains.go

@@ -0,0 +1,84 @@
+package validators
+
+import (
+	"fmt"
+
+	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
+	frameworkvalidators "gitea.maximumdirect.net/eric/audita/internal/framework/validators"
+)
+
+var builtInChains = map[string][]string{
+	"glossary": {
+		KeyNoEffect,
+		KeyOriginalTextPresence,
+		KeyConfidenceThreshold,
+		KeyProtectedTerms,
+		KeyNonEmptyCorrectedText,
+		KeySpokenFormPlausibility,
+		KeyMeaningReversalReview,
+	},
+	"homophones": {
+		KeyNoEffect,
+		KeyOriginalTextPresence,
+		KeyConfidenceThreshold,
+		KeyProtectedTerms,
+		KeyNonEmptyCorrectedText,
+		KeySpokenFormPlausibility,
+		KeyMeaningReversalReview,
+	},
+	"spoken_word": {
+		KeyNoEffect,
+		KeyOriginalTextPresence,
+		KeyConfidenceThreshold,
+		KeyProtectedTerms,
+		KeyNonEmptyCorrectedText,
+		KeySpokenWordReview,
+		KeyMeaningReversalReview,
+	},
+	"grammar": {
+		KeyNoEffect,
+		KeyOriginalTextPresence,
+		KeyConfidenceThreshold,
+		KeyProtectedTerms,
+		KeyNonEmptyCorrectedText,
+		KeyGrammarReview,
+		KeyMeaningReversalReview,
+	},
+}
+
+func BuiltInChainKeys(moduleKey string) ([]string, error) {
+	keys, ok := builtInChains[moduleKey]
+	if !ok {
+		return nil, fmt.Errorf("no built-in validator chain for module %q", moduleKey)
+	}
+	out := make([]string, len(keys))
+	copy(out, keys)
+	return out, nil
+}
+
+func ResolveBuiltInChain(moduleKey string, registry *Registry) ([]contracts.Validator, error) {
+	keys, err := BuiltInChainKeys(moduleKey)
+	if err != nil {
+		return nil, err
+	}
+	if registry == nil {
+		registry = NewBuiltInRegistry()
+	}
+
+	out := make([]contracts.Validator, 0, len(keys))
+	for _, key := range keys {
+		if moduleKey == "glossary" && key == KeyProtectedTerms {
+			// Glossary stages preserve current stricter protection semantics while
+			// reporting the stable protected_terms key.
+			v := frameworkvalidators.GlossaryStageProtectedGlossaryTermValidator{}
+			out = append(out, v)
+			continue
+		}
+		v, buildErr := registry.MustBuild(key)
+		if buildErr != nil {
+			return nil, buildErr
+		}
+		out = append(out, v)
+	}
+	return out, nil
+}

internal/validators/module_chains_test.go

@@ -0,0 +1,66 @@
+package validators_test
+
+import (
+	"testing"
+
+	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/audita/internal/modules/glossary"
+	"gitea.maximumdirect.net/eric/audita/internal/modules/grammar"
+	"gitea.maximumdirect.net/eric/audita/internal/modules/homophones"
+	"gitea.maximumdirect.net/eric/audita/internal/modules/spoken_word"
+	builtinvalidators "gitea.maximumdirect.net/eric/audita/internal/validators"
+)
+
+func TestProductionModulesUseRegisteredValidatorKeys(t *testing.T) {
+	registry := builtinvalidators.NewBuiltInRegistry()
+	keys := map[string]struct{}{}
+	for _, key := range registry.RegisteredKeys() {
+		keys[key] = struct{}{}
+	}
+
+	cases := []struct {
+		name  string
+		build func() ([]contracts.Validator, error)
+	}{
+		{name: "glossary", build: func() ([]contracts.Validator, error) {
+			m, err := glossary.New()
+			if err != nil {
+				return nil, err
+			}
+			return m.Validators(), nil
+		}},
+		{name: "homophones", build: func() ([]contracts.Validator, error) {
+			m, err := homophones.New()
+			if err != nil {
+				return nil, err
+			}
+			return m.Validators(), nil
+		}},
+		{name: "spoken_word", build: func() ([]contracts.Validator, error) {
+			m, err := spoken_word.New()
+			if err != nil {
+				return nil, err
+			}
+			return m.Validators(), nil
+		}},
+		{name: "grammar", build: func() ([]contracts.Validator, error) {
+			m, err := grammar.New()
+			if err != nil {
+				return nil, err
+			}
+			return m.Validators(), nil
+		}},
+	}
+
+	for _, tc := range cases {
+		validators, err := tc.build()
+		if err != nil {
+			t.Fatalf("build module %s: %v", tc.name, err)
+		}
+		for _, v := range validators {
+			if _, ok := keys[v.Name()]; !ok {
+				t.Fatalf("module %s uses unregistered validator key %q", tc.name, v.Name())
+			}
+		}
+	}
+}

internal/validators/registry.go

@@ -0,0 +1,104 @@
+package validators
+
+import (
+	"fmt"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/audita/internal/framework/contracts"
+	frameworkvalidators "gitea.maximumdirect.net/eric/audita/internal/framework/validators"
+)
+
+const (
+	KeyConfidenceThreshold   = "confidence_threshold"
+	KeyOriginalTextPresence  = "original_text_presence"
+	KeyNonEmptyCorrectedText = "non_empty_corrected_text"
+	KeyNoEffect              = "no_effect"
+	KeyProtectedTerms        = "protected_terms"
+
+	KeySpokenFormPlausibility = "spoken_form_plausibility"
+	KeyMeaningReversalReview  = "meaning_reversal_review"
+	KeyEditorialReview        = "editorial_review"
+	KeyGrammarReview          = "grammar_review"
+	KeySpokenWordReview       = "spoken_word_review"
+)
+
+type BuiltInValidatorDefinition struct {
+	Key       string
+	Build     func() (contracts.Validator, error)
+	LLMBacked bool
+}
+
+type Registry struct {
+	definitions map[string]BuiltInValidatorDefinition
+}
+
+func NewBuiltInRegistry() *Registry {
+	defs := []BuiltInValidatorDefinition{
+		{Key: KeyConfidenceThreshold, Build: func() (contracts.Validator, error) { return frameworkvalidators.ConfidenceThresholdValidator{}, nil }},
+		{Key: KeyOriginalTextPresence, Build: func() (contracts.Validator, error) { return frameworkvalidators.OriginalTextPresenceValidator{}, nil }},
+		{Key: KeyNonEmptyCorrectedText, Build: func() (contracts.Validator, error) { return frameworkvalidators.NonEmptyCorrectionValidator{}, nil }},
+		{Key: KeyNoEffect, Build: func() (contracts.Validator, error) { return frameworkvalidators.NoEffectValidator{}, nil }},
+		{Key: KeyProtectedTerms, Build: func() (contracts.Validator, error) { return frameworkvalidators.ProtectedGlossaryTermValidator{}, nil }},
+		{Key: KeySpokenFormPlausibility, LLMBacked: true, Build: func() (contracts.Validator, error) {
+			return frameworkvalidators.NewLLMBackedValidator(KeySpokenFormPlausibility, frameworkvalidators.LLMValidatorTypeSpokenFormPlausibility, "")
+		}},
+		{Key: KeyMeaningReversalReview, LLMBacked: true, Build: func() (contracts.Validator, error) {
+			return frameworkvalidators.NewLLMBackedValidator(KeyMeaningReversalReview, frameworkvalidators.LLMValidatorTypeMeaningReversal, "")
+		}},
+		{Key: KeyEditorialReview, LLMBacked: true, Build: func() (contracts.Validator, error) {
+			return frameworkvalidators.NewLLMBackedValidator(KeyEditorialReview, frameworkvalidators.LLMValidatorTypeEditorialReview, "")
+		}},
+		{Key: KeyGrammarReview, LLMBacked: true, Build: func() (contracts.Validator, error) {
+			return frameworkvalidators.NewLLMBackedValidator(KeyGrammarReview, frameworkvalidators.LLMValidatorTypeGrammarReview, "")
+		}},
+		{Key: KeySpokenWordReview, LLMBacked: true, Build: func() (contracts.Validator, error) {
+			return frameworkvalidators.NewLLMBackedValidator(KeySpokenWordReview, frameworkvalidators.LLMValidatorTypeSpokenWordReview, "")
+		}},
+	}
+
+	m := make(map[string]BuiltInValidatorDefinition, len(defs))
+	for _, def := range defs {
+		m[def.Key] = def
+	}
+	return &Registry{definitions: m}
+}
+
+func (r *Registry) Lookup(key string) (BuiltInValidatorDefinition, bool) {
+	if r == nil {
+		return BuiltInValidatorDefinition{}, false
+	}
+	def, ok := r.definitions[strings.TrimSpace(key)]
+	return def, ok
+}
+
+func (r *Registry) MustBuild(key string) (contracts.Validator, error) {
+	if r == nil {
+		return nil, fmt.Errorf("validator registry is nil")
+	}
+	def, ok := r.Lookup(key)
+	if !ok {
+		return nil, fmt.Errorf("unknown validator key %q", key)
+	}
+	v, err := def.Build()
+	if err != nil {
+		return nil, fmt.Errorf("build validator %q: %w", key, err)
+	}
+	if v == nil {
+		return nil, fmt.Errorf("build validator %q: returned nil validator", key)
+	}
+	if v.Name() != def.Key {
+		return nil, fmt.Errorf("validator key/name mismatch for %q: got %q", key, v.Name())
+	}
+	return v, nil
+}
+
+func (r *Registry) RegisteredKeys() []string {
+	if r == nil {
+		return nil
+	}
+	keys := make([]string, 0, len(r.definitions))
+	for k := range r.definitions {
+		keys = append(keys, k)
+	}
+	return keys
+}

internal/validators/registry_test.go

@@ -0,0 +1,79 @@
+package validators
+
+import (
+	"testing"
+)
+
+func TestBuiltInRegistryRegistersAllKeys(t *testing.T) {
+	r := NewBuiltInRegistry()
+	for _, key := range []string{
+		KeyConfidenceThreshold,
+		KeyOriginalTextPresence,
+		KeyNonEmptyCorrectedText,
+		KeyNoEffect,
+		KeyProtectedTerms,
+		KeySpokenFormPlausibility,
+		KeyMeaningReversalReview,
+		KeyEditorialReview,
+		KeyGrammarReview,
+		KeySpokenWordReview,
+	} {
+		def, ok := r.Lookup(key)
+		if !ok {
+			t.Fatalf("expected validator key %q to be registered", key)
+		}
+		v, err := def.Build()
+		if err != nil {
+			t.Fatalf("build validator %q: %v", key, err)
+		}
+		if v.Name() != key {
+			t.Fatalf("expected validator name %q, got %q", key, v.Name())
+		}
+	}
+}
+
+func TestBuiltInRegistryUnknownKeyFails(t *testing.T) {
+	r := NewBuiltInRegistry()
+	if _, err := r.MustBuild("missing"); err == nil {
+		t.Fatalf("expected unknown validator key error")
+	}
+}
+
+func TestBuiltInChainKeysResolveForProductionModules(t *testing.T) {
+	for _, moduleKey := range []string{"glossary", "homophones", "spoken_word", "grammar"} {
+		keys, err := BuiltInChainKeys(moduleKey)
+		if err != nil {
+			t.Fatalf("resolve keys for %q: %v", moduleKey, err)
+		}
+		if len(keys) == 0 {
+			t.Fatalf("expected non-empty chain for %q", moduleKey)
+		}
+	}
+}
+
+func TestResolveBuiltInChainUsesRegisteredKeys(t *testing.T) {
+	r := NewBuiltInRegistry()
+	for _, moduleKey := range []string{"glossary", "homophones", "spoken_word", "grammar"} {
+		chain, err := ResolveBuiltInChain(moduleKey, r)
+		if err != nil {
+			t.Fatalf("resolve chain for %q: %v", moduleKey, err)
+		}
+		if len(chain) == 0 {
+			t.Fatalf("expected non-empty chain for %q", moduleKey)
+		}
+		for _, v := range chain {
+			if v == nil {
+				t.Fatalf("nil validator in %q chain", moduleKey)
+			}
+		}
+	}
+}
+
+func TestBuiltInChainUnknownModuleFails(t *testing.T) {
+	if _, err := BuiltInChainKeys("unknown"); err == nil {
+		t.Fatalf("expected unknown module chain failure")
+	}
+	if _, err := ResolveBuiltInChain("unknown", NewBuiltInRegistry()); err == nil {
+		t.Fatalf("expected unknown module chain failure")
+	}
+}