nold-ai
diff --git a/‎.cursor/commands/specfact.01-import.md‎
Lines changed: 66 additions & 4 deletions b/‎.cursor/commands/specfact.01-import.md‎
Lines changed: 66 additions & 4 deletions
diff --git a/‎.cursor/commands/specfact.02-plan.md‎
Lines changed: 59 additions & 1 deletion b/‎.cursor/commands/specfact.02-plan.md‎
Lines changed: 59 additions & 1 deletion
@@ -16,25 +16,86 @@ Import codebase → plan bundle. CLI extracts routes/schemas/relationships/contr
 
 **Target/Input**: `--bundle NAME` (optional, defaults to active plan), `--repo PATH`, `--entry-point PATH`, `--enrichment PATH`  
 **Output/Results**: `--report PATH`  
-**Behavior/Options**: `--shadow-only`, `--enrich-for-speckit`  
+**Behavior/Options**: `--shadow-only`, `--enrich-for-speckit/--no-enrich-for-speckit` (default: enabled, uses PlanEnricher for consistent enrichment)  
 **Advanced/Configuration**: `--confidence FLOAT` (0.0-1.0), `--key-format FORMAT` (classname|sequential)
 
 ## Workflow
 
 1. **Execute CLI**: `specfact import from-code [<bundle>] --repo <path> [options]`
    - CLI extracts: routes (FastAPI/Flask/Django), schemas (Pydantic), relationships, contracts (OpenAPI scaffolds), source tracking
    - Uses active plan if bundle not specified
+   - **Auto-enrichment enabled by default**: Automatically enhances vague acceptance criteria, incomplete requirements, and generic tasks using PlanEnricher (same logic as `plan review --auto-enrich`)
+   - Use `--no-enrich-for-speckit` to disable auto-enrichment
 
 2. **LLM Enrichment** (if `--enrichment` provided):
    - Read `.specfact/projects/<bundle>/enrichment_context.md`
    - Enrich: business context, "why" reasoning, missing acceptance criteria
    - Validate: contracts vs code, feature/story alignment
+   - Save enrichment report to `.specfact/projects/<bundle-name>/reports/enrichment/` (bundle-specific, Phase 8.5, if created)
 
 3. **Present**: Bundle location, report path, summary (features/stories/contracts/relationships)
 
 ## CLI Enforcement
 
-**ALWAYS execute CLI first**. Never modify `.specfact/` directly. Use CLI output as grounding.
+**CRITICAL**: Always use SpecFact CLI commands. See [CLI Enforcement Rules](./shared/cli-enforcement.md) for details.
+
+**Rules:**
+
+- Execute CLI first - never create artifacts directly
+- Use `--no-interactive` flag in CI/CD environments
+- Never modify `.specfact/` directly
+- Use CLI output as grounding for validation
+- Code generation requires LLM (only via AI IDE slash prompts, not CLI-only)
+
+## Dual-Stack Workflow (Copilot Mode)
+
+When in copilot mode, follow this three-phase workflow:
+
+### Phase 1: CLI Grounding (REQUIRED)
+
+```bash
+# Execute CLI to get structured output
+specfact import from-code [<bundle>] --repo <path> --no-interactive
+```
+
+**Capture**:
+
+- CLI-generated artifacts (plan bundles, reports)
+- Metadata (timestamps, confidence scores)
+- Telemetry (execution time, file counts)
+
+### Phase 2: LLM Enrichment (OPTIONAL, Copilot Only)
+
+**Purpose**: Add semantic understanding to CLI output
+
+**What to do**:
+
+- Read CLI-generated artifacts (use file reading tools for display only)
+- Research codebase for additional context
+- Identify missing features/stories
+- Suggest confidence adjustments
+- Extract business context
+
+**What NOT to do**:
+
+- ❌ Create YAML/JSON artifacts directly
+- ❌ Modify CLI artifacts directly (use CLI commands to update)
+- ❌ Bypass CLI validation
+- ❌ Write to `.specfact/` folder directly (always use CLI)
+- ❌ Use direct file manipulation tools for writing (use CLI commands)
+
+**Output**: Generate enrichment report (Markdown) saved to `.specfact/projects/<bundle-name>/reports/enrichment/` (bundle-specific, Phase 8.5)
+
+### Phase 3: CLI Artifact Creation (REQUIRED)
+
+```bash
+# Use enrichment to update plan via CLI
+specfact import from-code [<bundle>] --repo <path> --enrichment <enrichment-report> --no-interactive
+```
+
+**Result**: Final artifacts are CLI-generated with validated enrichments
+
+**Note**: If code generation is needed, use the validation loop pattern (see [CLI Enforcement Rules](./shared/cli-enforcement.md#standard-validation-loop-pattern-for-llm-generated-code))
 
 ## Expected Output
 
@@ -44,8 +105,9 @@ Import codebase → plan bundle. CLI extracts routes/schemas/relationships/contr
 ## Common Patterns
 
 ```bash
-/specfact.01-import --repo .                    # Uses active plan
-/specfact.01-import --bundle legacy-api --repo .
+/specfact.01-import --repo .                    # Uses active plan, auto-enrichment enabled by default
+/specfact.01-import --bundle legacy-api --repo . # Auto-enrichment enabled
+/specfact.01-import --repo . --no-enrich-for-speckit  # Disable auto-enrichment
 /specfact.01-import --repo . --entry-point src/auth/
 /specfact.01-import --repo . --enrichment report.md
 ```
 
@@ -71,7 +71,65 @@ specfact plan update-story [--bundle <name>] --feature <feature-key> --key <stor
 
 **CRITICAL**: Always use SpecFact CLI commands. See [CLI Enforcement Rules](./shared/cli-enforcement.md) for details.
 
-**Rules:** Execute CLI first, use `--no-interactive` in CI/CD, never modify `.specfact/` directly, use CLI output as grounding.
+**Rules:**
+
+- Execute CLI first - never create artifacts directly
+- Use `--no-interactive` flag in CI/CD environments
+- Never modify `.specfact/` directly
+- Use CLI output as grounding for validation
+- Code generation requires LLM (only via AI IDE slash prompts, not CLI-only)
+
+## Dual-Stack Workflow (Copilot Mode)
+
+When in copilot mode, follow this three-phase workflow:
+
+### Phase 1: CLI Grounding (REQUIRED)
+
+```bash
+# Execute CLI to get structured output
+specfact plan <operation> [--bundle <name>] [options] --no-interactive
+```
+
+**Capture**:
+
+- CLI-generated artifacts (plan bundles, features, stories)
+- Metadata (timestamps, confidence scores)
+- Telemetry (execution time, file counts)
+
+### Phase 2: LLM Enrichment (OPTIONAL, Copilot Only)
+
+**Purpose**: Add semantic understanding to CLI output
+
+**What to do**:
+
+- Read CLI-generated artifacts (use file reading tools for display only)
+- Research codebase for additional context
+- Identify missing features/stories
+- Suggest confidence adjustments
+- Extract business context
+
+**What NOT to do**:
+
+- ❌ Create YAML/JSON artifacts directly
+- ❌ Modify CLI artifacts directly (use CLI commands to update)
+- ❌ Bypass CLI validation
+- ❌ Write to `.specfact/` folder directly (always use CLI)
+- ❌ Use direct file manipulation tools for writing (use CLI commands)
+
+**Output**: Generate enrichment report (Markdown) or use `--batch-updates` JSON/YAML file
+
+### Phase 3: CLI Artifact Creation (REQUIRED)
+
+```bash
+# Use enrichment to update plan via CLI
+specfact plan update-feature [--bundle <name>] --key <key> [options] --no-interactive
+# Or use batch updates:
+specfact plan update-feature [--bundle <name>] --batch-updates <updates.json> --no-interactive
+```
+
+**Result**: Final artifacts are CLI-generated with validated enrichments
+
+**Note**: If code generation is needed, use the validation loop pattern (see [CLI Enforcement Rules](./shared/cli-enforcement.md#standard-validation-loop-pattern-for-llm-generated-code))
 
 ## Expected Output