| name | ontology-phase-3-synthesize |
| description | Phase 3 of Ontology Builder Pipeline. Synthesizes analysis results into complete Domain Requirement Document (DRD). Use after Phase 2 analysis is complete to generate structured DRD. |
Phase 3: Synthesize DRD
Transform analysis results into a complete, well-structured Domain Requirement Document.
Trigger
Execute when:
- Phase 2 analysis-report.md is available
- DRD regeneration requested
Process
Step 1: Load Analysis Results
Read:
_output/_logs/analysis-report.md_input/project-context.md(for header info)
Step 2: Generate DRD Structure
Create: _output/00-drd/DRD-[module].md
Follow the template below exactly.
Step 3: Write Each Section
Section 1: Context
- Copy from project-context.md
- Add scope from analysis
Section 2: Domain Concepts
- Transform consolidated entities into concept descriptions
- Write in natural language, not YAML
- Include "key characteristics" for each
- Mark confidence inline
Section 3: Business Rules
- Organize rules by category
- Use consistent ID format
- Include conditions and actions
Section 4: Workflows
- Write full workflow descriptions
- Include main flow and alternates
- Reference entities by name
Section 5: Data Examples
- Generate realistic examples
- Show different scenarios
- Use YAML format for clarity
Section 6: Integration Points
- List external system integrations
- Specify data direction and purpose
Step 4: Add Metadata
- Confidence markers throughout
- Source attributions
- Assumptions section
- Open questions section
Step 5: Self-Review
Before finalizing:
- All analysis entities covered?
- All workflows included?
- Business rules linked to entities/workflows?
- Examples are realistic?
- No TODO placeholders?
DRD Template
# Domain Requirement Document: [Module Name]
**Module**: [MODULE-CODE]
**Version**: 1.0
**Generated**: [timestamp]
**Generator**: AI Ontology Builder Pipeline
**Status**: Draft - Pending Review
---
## Document Information
| Field | Value |
|-------|-------|
| Project | [Project name] |
| Domain | [Domain type] |
| Region | [Region/Country] |
| Confidence Level | [Overall: HIGH/MEDIUM/LOW] |
| Assumptions Made | [Count] |
| Open Questions | [Count] |
---
## 1. Context
### 1.1 Business Background
[2-3 paragraphs describing business context, synthesized from inputs]
[CONFIDENCE: HIGH/MEDIUM/LOW] (Source: [sources])
### 1.2 Scope
**In Scope**:
- [Feature 1] [CONFIDENCE]
- [Feature 2] [CONFIDENCE]
- [Feature 3] [CONFIDENCE]
**Out of Scope**:
- [Item 1] → [Reason/Where handled]
- [Item 2] → [Reason]
### 1.3 Users & Roles
| Role | Description | Key Actions | Source |
|------|-------------|-------------|--------|
| [Role 1] | [Desc] | [Actions] | [Source] |
| [Role 2] | [Desc] | [Actions] | [Source] |
---
## 2. Domain Concepts
> These concepts form the foundation of the [Module] domain model.
### 2.1 Core Concepts
#### [Concept Name 1] [CONFIDENCE]
[2-3 sentence description in natural language]
**Key Characteristics**:
- [Characteristic 1]
- [Characteristic 2]
- [Characteristic 3]
**Attributes** (for reference):
- `[attribute_1]`: [type] - [description]
- `[attribute_2]`: [type] - [description]
**Related To**: [Entity1], [Entity2]
(Source: [source files] | [If ASSUMED: "Market knowledge: [reference]"])
---
#### [Concept Name 2] [CONFIDENCE]
[Description]
**Key Characteristics**:
- [Characteristic 1]
- [Characteristic 2]
...
### 2.2 Supporting Concepts
#### [Supporting Concept 1] [CONFIDENCE]
[Description]
---
## 3. Business Rules
### 3.1 [Category] Rules
| ID | Rule | Condition | Action | Source |
|----|------|-----------|--------|--------|
| BR-[MOD]-001 | [Short name] | [When condition] | [Then action] | [Source] |
| BR-[MOD]-002 | [Short name] | [When condition] | [Then action] | [Source] |
### 3.2 [Category] Rules
| ID | Rule | Condition | Action | Source |
|----|------|-----------|--------|--------|
| BR-[MOD]-010 | [Short name] | [When condition] | [Then action] | [Source] |
### 3.3 Assumed Rules (Market Standard)
> These rules are assumed based on industry standards and market knowledge.
| ID | Rule | Condition | Action | Reference |
|----|------|-----------|--------|-----------|
| BR-[MOD]-A01 | [Short name] | [When condition] | [Then action] | [Market reference] |
---
## 4. Workflows
### 4.1 WF-[MOD]-001: [Workflow Name] [CONFIDENCE]
**Classification**: [CORE | SUPPORT | INTEGRATION]
**Trigger**: [What initiates this workflow]
**Actors**: [Actor 1], [Actor 2], [System]
**Result**: [What is achieved when workflow completes]
**Main Flow**:
1. [Actor] [action description]
2. System [validation/processing description]
3. [Actor] [next action]
4. System [updates/notifications]
5. [Continue as needed...]
**Alternate Flows**:
- **[Step]a. [Condition]**: [What happens instead]
- **[Step]b. [Condition]**: [What happens instead]
**Exception Handling**:
| Exception | Trigger | System Response |
|-----------|---------|-----------------|
| [Exception 1] | [Condition] | [Response] |
| [Exception 2] | [Condition] | [Response] |
**Business Rules Applied**: BR-[MOD]-001, BR-[MOD]-002
**Related Entities**: [Entity1], [Entity2]
(Source: [sources])
---
### 4.2 WF-[MOD]-002: [Workflow Name] [CONFIDENCE]
[Same structure as above]
---
## 5. Data Examples
### 5.1 [Concept] Examples
```yaml
Example 1: [Scenario name]
[concept_name]:
[attribute_1]: [value]
[attribute_2]: [value]
[attribute_3]: [value]
Example 2: [Different scenario]
[concept_name]:
[attribute_1]: [different value]
[attribute_2]: [different value]
5.2 [Workflow] Scenario
Scenario: [Scenario description]
Initial State:
[entity]:
[attr]: [value]
status: [initial status]
Action: [What happens]
Final State:
[entity]:
[attr]: [new value]
status: [final status]
Side Effects:
- [Side effect 1]
- [Side effect 2]
6. Integration Points
| System | Direction | Data | Purpose | Priority |
|---|---|---|---|---|
| [System 1] | Inbound | [Data type] | [Purpose] | [HIGH/MED/LOW] |
| [System 2] | Outbound | [Data type] | [Purpose] | [Priority] |
| [System 3] | Bidirectional | [Data type] | [Purpose] | [Priority] |
7. Non-Functional Requirements
7.1 Performance
- [Requirement 1]
- [Requirement 2]
7.2 Security
- [Requirement 1]
- [Requirement 2]
7.3 Audit
- [Requirement 1]
8. Assumptions
These assumptions were made during analysis. Please validate.
| # | Assumption | Rationale | Impact if Wrong |
|---|---|---|---|
| A1 | [Assumption] | [Why assumed] | [Impact] |
| A2 | [Assumption] | [Why assumed] | [Impact] |
| A3 | [Assumption] | [Why assumed] | [Impact] |
9. Open Questions
These questions need stakeholder input for complete accuracy.
| # | Question | Context | Default if No Answer |
|---|---|---|---|
| Q1 | [Question] | [Why important] | [What we'll assume] |
| Q2 | [Question] | [Why important] | [What we'll assume] |
10. Confidence Summary
By Section
| Section | Confidence | Notes |
|---|---|---|
| Context | [HIGH/MED/LOW] | [Notes] |
| Domain Concepts | [%H/%M/%L] | [Notes] |
| Business Rules | [%H/%M/%L] | [Notes] |
| Workflows | [%H/%M/%L] | [Notes] |
By Source
| Source Type | Items | Confidence |
|---|---|---|
| User Stories | [N] | HIGH |
| Interviews | [N] | HIGH |
| Existing Docs | [N] | MEDIUM |
| Market Knowledge | [N] | ASSUMED |
11. References
Input Sources
- [List all input files used]
Market References
- [List industry standards/systems referenced]
Next Steps
- Review assumptions (Section 8)
- Answer open questions (Section 9)
- Validate business rules (Section 3)
- Approve for Ontology generation
Document History
| Version | Date | Changes |
|---|---|---|
| 1.0 | [date] | Initial generation by AI Pipeline |
## Output
- `_output/00-drd/DRD-[module].md`
- `_output/_logs/gate-3-manifest.yaml` (verification manifest)
- Ready for Phase 4: Generate Ontology
## Gate 3: Self-Verification
Before completing Phase 3, generate verification manifest:
```yaml
# _output/_logs/gate-3-manifest.yaml
gate: 3
name: "Post-DRD Verification"
timestamp: "[ISO timestamp]"
structural_checks:
- check: "DRD document exists"
status: PASS
- check: "All required sections present"
status: PASS | FAIL
sections:
context: true | false
domain_concepts: true | false
business_rules: true | false
workflows: true | false
examples: true | false
assumptions: true | false
- check: "No TODO/TBD placeholders"
status: PASS | FAIL
placeholders_found: []
consistency_checks:
- check: "All analysis entities in DRD"
status: PASS | FAIL
analysis_count: [N]
drd_count: [N]
missing: []
- check: "All analysis workflows in DRD"
status: PASS | FAIL
missing: []
- check: "Business rule IDs unique"
status: PASS | FAIL
duplicates: []
- check: "Workflow IDs unique"
status: PASS | FAIL
duplicates: []
traceability_checks:
- check: "Concepts trace to analysis"
status: PASS | FAIL
- check: "Confidence markers present"
status: PASS | FAIL
sections_without_confidence: []
quality_checks:
- check: "Business rules have condition and action"
status: PASS | FAIL
vague_rules: []
- check: "Workflows have complete main flow"
status: PASS | FAIL
incomplete_workflows: []
- check: "Examples are realistic"
status: PASS | FAIL
result:
status: PASS | FAIL | WARN
blocking_failures: []
warnings: []
proceed_to_next_phase: true | false
Verification Rules:
- FAIL if missing required sections
- FAIL if TODO/TBD placeholders remain
- WARN if any analysis entity missing from DRD
Quality Checklist
Before proceeding to Phase 4:
- All consolidated entities from analysis appear in Section 2
- All workflows from analysis appear in Section 4
- All business rules have IDs and are linked
- Examples are realistic and consistent
- Assumptions clearly documented
- Confidence markers throughout
Next Phase
After completing DRD:
→ Load phase-4-generate/SKILL.md
→ Pass DRD as input