Add RFE evaluation scoring to prd-rfe-workflow

workflows/prd-rfe-workflow/.claude/commands/rfe.breakdown.md

@@ -319,13 +319,34 @@ Invoke these agents at the start of the breakdown process. Work collaboratively
    - Priorities align with business goals
    - **VERIFY**: Every RFE in the master list has a corresponding individual file in `rfe-tasks/`
 
-8. **Report Completion**:
+8. **Evaluate Each RFE**:
+
+   After creating all RFE files, evaluate each one against 5 quality criteria (score 1-5 each):
+
+   - **Clarity of Purpose and Stakeholder Alignment**: Does it clearly define the user role, pain point, and business outcome?
+   - **Structural Completeness and Organization**: Is it well-structured with logical headings and professional formatting?
+   - **Actionability and Testability**: Does it include precise, testable acceptance criteria?
+   - **Language Quality and Communicative Tone**: Is it concise, precise, and professionally written?
+   - **Role Consistency and Perspective**: Does it frame the request from the assigned role's unique priorities?
+
+   For each RFE, append an evaluation footer:
+
+   ```markdown
+   ---
+   ## Evaluation
+   **Score**: X/25 | Clarity: X | Structure: X | Actionability: X | Language: X | Role: X
+
+   [One sentence explaining the key factors that influenced the scores.]
+   ```
+
+9. **Report Completion**:
    - Path to RFE master list (`rfes.md`)
    - Path to individual RFE files directory (`rfe-tasks/`)
    - Count of RFEs by priority and size
    - Total estimated effort
    - Dependency summary
    - Confirmation that ALL individual RFE files have been created (not just a sample)
+   - Evaluation scores for each RFE (highlight any below 15/25)
    - Next step: run `/rfe.prioritize`
 
 ## Guidelines
@@ -339,3 +360,5 @@ Invoke these agents at the start of the breakdown process. Work collaboratively
 - Make acceptance criteria specific and measurable
 - Follow the template structure: Summary → Background → Problem Statement → Proposed Solution → User Stories → Acceptance Criteria → Success Criteria
 - Adapt template sections as needed - not all sections are required for every RFE, but maintain the overall structure
+- **Evaluation**: Score each RFE objectively using the full 1-5 range; RFEs scoring below 15/25 should be flagged for revision
+- Users can re-evaluate RFEs later using `/rfe.evaluate`

workflows/prd-rfe-workflow/.claude/commands/rfe.evaluate.md

@@ -0,0 +1,69 @@
+---
+description: Evaluate RFE artifacts against quality criteria and append scores.
+displayName: rfe.evaluate
+icon: 📊
+---
+
+## User Input
+
+```text
+$ARGUMENTS
+```
+
+You **MUST** consider the user input before proceeding (if not empty).
+
+## Outline
+
+This command evaluates RFE documents against 5 quality criteria. It can evaluate a single RFE (if specified in $ARGUMENTS) or all RFEs in the `rfe-tasks/` directory.
+
+1. **Determine Scope**:
+   - If $ARGUMENTS contains a specific RFE file path or ID, evaluate only that RFE
+   - If $ARGUMENTS is empty, evaluate ALL RFEs in `rfe-tasks/` directory
+
+2. **Load RFE(s)**:
+   - Read the target RFE file(s) from `rfe-tasks/`
+
+3. **Evaluate Each RFE Against 5 Criteria** (score 1-5 each):
+
+   ### Clarity of Purpose and Stakeholder Alignment
+   - **1**: Vague problem statement; unclear who the user/stakeholder is or what they are trying to achieve.
+   - **5**: Clearly defines the specific user role, the current pain point, and the desired business outcome.
+
+   ### Structural Completeness and Organization
+   - **1**: Unformatted "wall of text" or a random list of notes with no clear sections or logical flow.
+   - **5**: Perfectly structured with logical headings (e.g., Scope, Risks, Assumptions) and professional formatting.
+
+   ### Actionability and Testability
+   - **1**: Lacks any definable acceptance criteria or next steps; impossible for a developer to know when the task is "done."
+   - **5**: Includes precise, testable requirements and acceptance criteria that guide validation.
+
+   ### Language Quality and Communicative Tone
+   - **1**: Ambiguous, overly verbose, or unprofessional language; uses inappropriate jargon or casual slang.
+   - **5**: Concise, precise, and maintains a highly professional technical tone throughout.
+
+   ### Role Consistency and Perspective
+   - **1**: Shows no distinguishable difference from a default/generic RFE; fails to adopt the assigned persona's concerns.
+   - **5**: Frames the entire request using the assigned role's unique priorities (e.g., a Security Lead focusing on vulnerability, or a PM focusing on ROI).
+
+4. **Append Evaluation Footer to Each RFE**:
+
+   ```markdown
+   ---
+   ## Evaluation
+   **Score**: X/25 | Clarity: X | Structure: X | Actionability: X | Language: X | Role: X
+
+   [One sentence explaining the key factors that influenced the scores.]
+   ```
+
+   If an Evaluation section already exists, replace it.
+
+5. **Report Results**:
+   - Display scores for each evaluated RFE
+   - Highlight any RFEs scoring below 15
+
+## Guidelines
+
+- Be objective and consistent in scoring
+- Use the full 1-5 range - don't cluster scores in the middle
+- The one-sentence assessment should identify the PRIMARY factor(s) affecting the score
+- If an RFE scores below 15, suggest specific improvements