speckeeper 0.10.1 → 0.11.0

package/docs/design/functional-requirements.md

@@ -51,6 +51,11 @@
 | FR-1017 | Source Path Fallback | must | check |
 | FR-1018 | Minimal New Dependencies | must | check |
 | FR-1019 | Checker Documentation Accuracy | must | check |
+| FR-1100 | LLM-Powered Requirement Audit | should | llm |
+| FR-1101 | LLM-Powered Trace Link Proposal | should | llm |
+| FR-1102 | LLM-Powered Impact Explanation | should | llm |
+| FR-1103 | LLM-Powered Acceptance Criteria Proposal | should | llm |
+| FR-1104 | TypeScript to YAML Conversion | should | conversion |
 
 ---
 
@@ -395,8 +400,8 @@ Since consistency checks are implemented per model, filter by model name
 ### Acceptance Criteria
 
 - **FR-602-01**: speckeeper check runs external SSOT consistency check for all models [test]
-- **FR-602-02**: speckeeper check --model <model-name> checks only specific model [test]
-- **FR-602-03**: Model name is the model ID defined in design/_models/ [review]
+- **FR-602-02**: speckeeper check [type] filters checks by type (openapi, ddl, iac, external-ssot, test, contract) [test]
+- **FR-602-03**: Type argument corresponds to check categories, not model IDs [review]
 - **FR-602-04**: Only models with externalChecker are targeted [test]
 
 ---
@@ -807,4 +812,97 @@ README and scaffold-mermaid-spec.md accurately describe all three built-in check
 ### Acceptance Criteria
 
 - **FR-1019-01**: README checker table describes validation levels [review]
-- **FR-1019-02**: scaffold-mermaid-spec.md Section 7 describes validation levels [review]
+- **FR-1019-02**: scaffold-mermaid-spec.md Section 7 describes validation levels [review]
+
+---
+
+## FR-1100: LLM-Powered Requirement Audit
+
+**Type**: functional | **Priority**: should | **Category**: llm
+
+Provide LLM-based semantic quality audit for requirement definitions, detecting verifiability issues, ambiguity, granularity problems, terminology inconsistency, and design-mixing
+
+### Rationale
+
+Static lint rules cannot detect semantic issues such as ambiguous wording or design details mixed into requirements; LLM review complements structural checks
+
+### Acceptance Criteria
+
+- **FR-1100-01**: audit-requirements command constructs a prompt from all registered specs and sends it to configured LLM adapter [test]
+- **FR-1100-02**: Audit report includes findings with severity (error/warning/info) and affected spec IDs [test]
+- **FR-1100-03**: --dry-run outputs the constructed prompt without calling LLM [test]
+- **FR-1100-04**: --fail-on controls minimum severity that causes non-zero exit [test]
+- **FR-1100-05**: Report format is selectable via --report-format (json, text, yaml) [test]
+
+---
+
+## FR-1101: LLM-Powered Trace Link Proposal
+
+**Type**: functional | **Priority**: should | **Category**: llm
+
+Propose candidate traceability links between specs with confidence scores and rationale using LLM analysis
+
+### Rationale
+
+Manual traceability maintenance is error-prone; LLM can identify semantically related specs that humans may overlook
+
+### Acceptance Criteria
+
+- **FR-1101-01**: propose-trace-links command analyzes all specs and proposes missing trace links [test]
+- **FR-1101-02**: Each proposed link includes source ID, target ID, relation type, confidence score, and rationale [test]
+- **FR-1101-03**: --dry-run outputs the constructed prompt without calling LLM [test]
+
+---
+
+## FR-1102: LLM-Powered Impact Explanation
+
+**Type**: functional | **Priority**: should | **Category**: llm
+
+Translate impact analysis JSON output into human-readable explanation for PM/executive audiences using LLM
+
+### Rationale
+
+Raw impact analysis JSON is not consumable by non-technical stakeholders; LLM can generate natural language summaries
+
+### Acceptance Criteria
+
+- **FR-1102-01**: explain-impact command reads impact analysis JSON from stdin [test]
+- **FR-1102-02**: Output is a human-readable explanation suitable for PM/executive audiences [review]
+- **FR-1102-03**: --dry-run outputs the constructed prompt without calling LLM [test]
+
+---
+
+## FR-1103: LLM-Powered Acceptance Criteria Proposal
+
+**Type**: functional | **Priority**: should | **Category**: llm
+
+Propose testable acceptance criteria in Given/When/Then format for specified specs using LLM
+
+### Rationale
+
+Writing testable acceptance criteria is time-consuming; LLM can propose initial criteria that humans refine
+
+### Acceptance Criteria
+
+- **FR-1103-01**: propose-acceptance-criteria command generates criteria for specified spec IDs (or all) [test]
+- **FR-1103-02**: Proposed criteria follow Given/When/Then format [review]
+- **FR-1103-03**: --dry-run outputs the constructed prompt without calling LLM [test]
+
+---
+
+## FR-1104: TypeScript to YAML Conversion
+
+**Type**: functional | **Priority**: should | **Category**: conversion
+
+Convert TypeScript spec data files to equivalent YAML format, enabling non-TypeScript workflows and interoperability
+
+### Rationale
+
+YAML input lowers participation barriers for non-developers (NFR-005) and enables interoperability with external tools
+
+### Acceptance Criteria
+
+- **FR-1104-01**: convert command reads a TS file exporting a SpecModule via defineSpecs() and writes equivalent YAML [test]
+- **FR-1104-02**: Output defaults to same filename with .yaml extension [test]
+- **FR-1104-03**: --output allows specifying a custom output path [test]
+- **FR-1104-04**: --dry-run previews conversion without writing files [test]

package/docs/design/nonfunctional-requirements.md

@@ -214,15 +214,15 @@ To ensure all acceptance criteria are covered by test cases and maintain spec-te
 
 **Type**: non-functional | **Priority**: must | **Category**: testability
 
-CLI command definitions in design/cli-commands.ts match actual implementation in src/cli/index.ts
+CLI command definitions in design/cli-commands.ts match cli-contract.yaml and generated code in src/generated/
 
 ### Rationale
 
-To ensure specification and implementation stay synchronized (e.g., no missing --config parameters)
+To ensure specification, DSL contract, and generated implementation stay synchronized
 
 ### Acceptance Criteria
 
-- **NFR-014-01**: All command definitions in design/cli-commands.ts match implementation (parameters, subcommands, exit codes) [test]
+- **NFR-014-01**: All command definitions in design/cli-commands.ts match cli-contract.yaml and generated code (parameters, subcommands, exit codes) [test]
 
 ---
 

package/docs/design/test-refs.md

@@ -12,6 +12,7 @@
 | TEST-023 | Impact command verification test | vitest | 1 |
 | TEST-024 | Drift command verification test | vitest | 1 |
 | TEST-025 | New command verification test | vitest | 1 |
+| TEST-026 | Scaffold integration verification test (mermaid parsing, class-based generation) | vitest | 1 |
 
 ---
 
@@ -256,3 +257,28 @@
 | FR-104-01 | `FR-104-01.*available model types header` | Outputs model types header when type omitted |
 
 ---
+
+## TEST-026: Scaffold integration verification test (mermaid parsing, class-based generation)
+
+### Test Source
+
+- **Path**: `test/scaffold/integration.test.ts`
+- **Framework**: vitest
+
+### Verified Requirements
+
+- FR-106
+
+### Implemented Command
+
+- CMD-SCAFFOLD
+
+### Test Case Patterns
+
+| Acceptance Criteria ID | Pattern | Description |
+|------------------------|---------|-------------|
+| FR-106-01 | `base template.*core factory|generated models.*base template` | Artifact class generates from base template |
+| FR-106-03 | `SR.*FR.*NFR.*map to requirement.*de-duplicated` | Same-class node aggregation into single model file |
+| FR-106-05 | `de-duplicated model files.*spec data` | Model file generation with naming conventions |
+
+---

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "speckeeper",
-  "version": "0.10.1",
+  "version": "0.11.0",
   "description": "TypeScript-first specification validation framework with external SSOT integration",
   "type": "module",
   "main": "./dist/index.js",
@@ -63,6 +63,7 @@
   "license": "MIT",
   "dependencies": {
     "chalk": "^5.3.0",
+    "cli-contracts": "^0.7.0",
     "commander": "^12.1.0",
     "embedoc": "^0.10.1",
     "glob": "^10.3.10",
@@ -86,9 +87,8 @@
     "@typescript-eslint/eslint-plugin": "^8.54.0",
     "@typescript-eslint/parser": "^8.54.0",
     "@vitest/coverage-v8": "^1.6.1",
-    "agent-contracts": "^0.21.0",
+    "agent-contracts": "^0.22.3",
     "agent-contracts-runtime": "^0.13.0",
-    "cli-contracts": "^0.6.1",
     "eslint": "^9.39.2",
     "tsup": "^8.0.1",
     "typescript": "^5.3.3",