name: {guru-id} description: {Brief description for Claude Code skill invocation. Include primary triggers and capabilities.}

Optional: Document common aliases (NOT a supported feature - documentation only)

Common short forms: {alias1}, {alias2} # Example: qa, tester

{Guru Name} Skill

You are a specialized {domain} agent for the morphir-dotnet project. Your role is to {primary responsibility} through {key capabilities}.

Primary Responsibilities

1. {Responsibility 1} - {Brief description}
2. {Responsibility 2} - {Brief description}
3. {Responsibility 3} - {Brief description}
4. {Responsibility 4} - {Brief description}
5. {Responsibility 5} - {Brief description}
6. {Responsibility 6} - {Brief description}

Core Competencies

{Competency 1}

When asked to {task related to competency 1}:

1. {Step 1}
2. {Step 2}
3. {Step 3}
4. {Step 4}
5. {Step 5}

Example:

{Code example showing competency 1}

{Competency 2}

When performing {task related to competency 2}:

1. {Step 1}
2. {Step 2}
3. {Step 3}
4. {Step 4}

Key Principles:

• {Principle 1}
• {Principle 2}
• {Principle 3}

{Competency 3}

When {scenario for competency 3}:

1. {Step 1}
2. {Step 2}
3. {Step 3}

Checklist:

• {Check 1}
• {Check 2}
• {Check 3}
• {Check 4}

{Competency 4}

{Description of competency 4}

Process:

{Visual representation of process}

Common Patterns:

• {Pattern Name 1}: {When to use} → {How to apply}
• {Pattern Name 2}: {When to use} → {How to apply}
• {Pattern Name 3}: {When to use} → {How to apply}

Project-Specific Context

morphir-dotnet {Domain} Specifics

{Project-specific information relevant to this guru's domain}

Key Areas:

1. {Area 1} - {Description}
2. {Area 2} - {Description}
3. {Area 3} - {Description}

Important Files and Directories

{Relevant file structure}

Commands and Tools

# {Command purpose 1}
{command1}

# {Command purpose 2}
{command2}

# {Command purpose 3}
{command3}

Decision Trees

Decision Tree 1: "{When to make decision 1}"

{Question 1}?
  YES → {Action or next question}
    ├─ {Sub-question 1}?
    │  ├─ YES → {Action}
    │  └─ NO → {Action}
    └─ {Sub-question 2}?
       ├─ YES → {Action}
       └─ NO → {Action}
  
  NO → {Alternative question or action}
    └─ {Follow-up}

Decision Tree 2: "{When to make decision 2}"

{Scenario description}?
  ├─ {Option A} → {Action A}
  ├─ {Option B} → {Action B}
  ├─ {Option C} → {Action C}
  └─ {Option D} → {Action D}

After decision:
  → {Follow-up step 1}
  → {Follow-up step 2}

Decision Tree 3: "{When to make decision 3}"

1. {First check}
   A. {Condition A}
      → {Action for A}
   
   B. {Condition B}
      → {Action for B}
   
   C. {Condition C}
      → {Action for C}

2. After resolution:
   → {Follow-up}
   → {Validation}

Playbooks

Playbook 1: {Workflow Name}

When to use: {Scenario description}

Prerequisites:

• {Prerequisite 1}
• {Prerequisite 2}
• {Prerequisite 3}

Steps:

Phase 1: {Phase Name}

1. {Step Name}

   {command if applicable}
   

   • Expected result: {What you should see}
   • If fails: {Troubleshooting}

2. {Step Name}

   • {Action to take}
   • Validation: {How to verify success}

3. {Step Name}

   • {Action to take}
   • Note: {Important information}

Phase 2: {Phase Name} 4. {Step Name}

{command if applicable}

• {Important detail}

5. {Step Name}
   • {Action to take}
   • Why: {Rationale}

Phase 3: {Phase Name} 6. {Step Name}

• {Action to take}
• Verification: {Validation method}

7. {Step Name}
   • {Final action}
   • Success criteria: {What success looks like}

Post-Workflow:

• {Cleanup task 1}
• {Documentation task}
• {Feedback capture}

Duration: ~{X} minutes

Playbook 2: {Workflow Name}

{Similar structure to Playbook 1}

Playbook 3: {Workflow Name}

{Similar structure to Playbook 1}

Review Capability

Review Scope

This guru proactively reviews the {domain area} for:

1. {Issue Category 1} - {What to look for}

   • Example: {Specific pattern or anti-pattern}
   • Detection: {How to identify}
   • Impact: {Why this matters}

2. {Issue Category 2} - {What to look for}

   • Example: {Specific pattern or anti-pattern}
   • Detection: {How to identify}
   • Impact: {Why this matters}

3. {Issue Category 3} - {What to look for}

   • Example: {Specific pattern or anti-pattern}
   • Detection: {How to identify}
   • Impact: {Why this matters}

4. {Issue Category 4} - {What to look for}

   • Example: {Specific pattern or anti-pattern}
   • Detection: {How to identify}
   • Impact: {Why this matters}

Review Triggers

Continuous Review (if applicable):

• Trigger: {When continuous review runs}
• Scope: {What gets reviewed}
• Output: {Where results go}

Scheduled Review (if applicable):

• Frequency: {How often} (e.g., quarterly, weekly)
• Trigger: {What initiates it}
• Scope: {What gets reviewed}
• Output: {Report format and location}

Session-Based Review (if applicable):

• Trigger: {After what workflow}
• Scope: {What gets reviewed}
• Output: {Integrated into workflow results}

On-Demand Review (if applicable):

• Trigger: Manual request (@guru review)
• Scope: Full domain scan
• Output: Comprehensive report

Review Output Format

Findings Structure:

# {Guru Name} Review Report
**Date:** {YYYY-MM-DD}
**Scope:** {What was reviewed}
**Duration:** {Review duration}

## Summary
- {Metric 1}: {Value}
- {Metric 2}: {Value}
- {Metric 3}: {Value}

## Findings

### Category: {Issue Category 1}
- **Finding 1:** {Issue description}
  - **Location:** {File/line/component}
  - **Severity:** {Critical/High/Medium/Low}
  - **Recommendation:** {Suggested fix}
  
- **Finding 2:** {Issue description}
  - **Location:** {File/line/component}
  - **Severity:** {Critical/High/Medium/Low}
  - **Recommendation:** {Suggested fix}

### Category: {Issue Category 2}
{Similar structure}

## Trends
- {Trend 1}: {Description and data}
- {Trend 2}: {Description and data}

## Recommendations
1. **Immediate:** {High-priority recommendations}
2. **Short-term:** {Next quarter priorities}
3. **Long-term:** {Future improvements}

## Automation Opportunities
- {Pattern 1}: Appears {N} times → {Suggested automation}
- {Pattern 2}: Appears {N} times → {Suggested automation}

## Next Review Focus
- {Area to monitor closely}
- {Threshold to watch}

Integration with Retrospectives

Reviews and retrospectives work together for continuous improvement:

Review (Proactive):

• "I scanned {domain} and found these issues"
• Identifies patterns, trends, violations
• Feeds findings to retrospectives

Retrospective (Reactive):

• "That failure happened because of X"
• Root cause analysis
• Feeds prevention strategies to reviews

Continuous Improvement Cycle:

Review Findings
    ↓
Identify Root Causes (Retrospective)
    ↓
Implement Prevention
    ↓
Update Playbooks/Automation
    ↓
Update Review Criteria
    ↓
Next Review (with better focus)

Example Integration:

Q1 Review: Found {N} instances of {anti-pattern}
Q1 Retrospective: Analyzed why {anti-pattern} occurred
Q1 Action: Created {automation} to detect/prevent
Q2 Review: {Anti-pattern} instances reduced to 0
Q2 Outcome: Pattern successfully eliminated

Review Automation Scripts

Location: .claude/skills/{guru-id}/scripts/

{review-script-1}.fsx

• Purpose: {What this review script does}
• Triggers: {When it runs}
• Output: {What it produces}
• Token Savings: ~{N} tokens (vs manual review)

{review-script-2}.fsx

• Purpose: {What this review script does}
• Triggers: {When it runs}
• Output: {What it produces}
• Token Savings: ~{N} tokens (vs manual review)

Usage:

# Run {review type 1}
dotnet fsi .claude/skills/{guru-id}/scripts/{review-script-1}.fsx

# Run {review type 2}
dotnet fsi .claude/skills/{guru-id}/scripts/{review-script-2}.fsx

Review Checklist

Before completing a review:

• All defined scope areas covered
• Findings categorized by severity
• Recommendations provided for each finding
• Trends analyzed (vs previous review)
• Automation opportunities identified
• Report generated and saved
• Key stakeholders notified (if applicable)
• Next review criteria updated
• Learnings fed to playbooks/decision trees

Pattern Catalog

Pattern 1: {Pattern Name}

Category: {Pattern category}
Frequency: {How often seen}
Complexity: {Low/Medium/High}

Problem: {What problem does this pattern solve or represent?}

Solution:

{Code example showing the pattern}

When to Use:

• {Scenario 1}
• {Scenario 2}
• {Scenario 3}

When to Avoid:

• {Anti-pattern scenario 1}
• {Anti-pattern scenario 2}

Related Patterns:

• {Related Pattern 1}
• {Related Pattern 2}

Pattern 2: {Pattern Name}

{Similar structure to Pattern 1}

Pattern 3: {Pattern Name}

{Similar structure to Pattern 1}

{Continue with more patterns...}

Automation Scripts

Location: .claude/skills/{guru-id}/scripts/

Script 1: {script-name-1}.fsx

Purpose: {What this script does}
Token Savings: ~{N} tokens per use (vs {manual alternative})

Usage:

dotnet fsi .claude/skills/{guru-id}/scripts/{script-name-1}.fsx [args]

Arguments:

• {arg1} - {Description}
• {arg2} - {Description}
• {arg3} - {Description} (optional)

Output: {Description of output format}

Example:

# {Example usage scenario}
dotnet fsi .claude/skills/{guru-id}/scripts/{script-name-1}.fsx --{arg1}=value

Script 2: {script-name-2}.fsx

{Similar structure to Script 1}

Script 3: {script-name-3}.fsx

{Similar structure to Script 1}

Integration Points

Coordination with Other Gurus

{Other Guru 1}:

• Direction: {to/from} this guru
• Interaction: {What gets passed}
• Trigger: {When this happens}
• Protocol: {How the hand-off works}

{Other Guru 2}:

• Direction: {to/from} this guru
• Interaction: {What gets passed}
• Trigger: {When this happens}
• Protocol: {How the hand-off works}

{Other Guru 3}:

• Direction: {to/from} this guru
• Interaction: {What gets passed}
• Trigger: {When this happens}
• Protocol: {How the hand-off works}

Escalation Paths

When to Escalate:

1. {Scenario requiring human decision}
2. {Scenario beyond scope}
3. {Scenario with high uncertainty}

How to Escalate:

1. Document the decision point
2. Provide context and options
3. Tag appropriate maintainer: {maintainer-tag}
4. Label issue: maintainer-attention
5. Wait for guidance

What NOT to Decide:

• {Out-of-scope decision 1}
• {Out-of-scope decision 2}
• {Out-of-scope decision 3}

Feedback Loop

Feedback Capture

Trigger Points:

• {When feedback is captured - e.g., "After each {workflow}"}
• {When feedback is captured - e.g., "Every {N} {units}"}
• {When feedback is captured - e.g., "Quarterly review"}

Capture Method: {How feedback is recorded - e.g., "Template section", "GitHub issue", "IMPLEMENTATION.md"}

What to Capture:

• Patterns discovered
• Decision points encountered
• Edge cases found
• Improvements identified
• Automation opportunities

Quarterly Review Process

Schedule: {When reviews happen - e.g., "Q1, Q2, Q3, Q4"}

Review Checklist:

1. Collect all feedback from quarter
2. Analyze patterns and trends
3. Identify 2-3 key improvements
4. Update playbooks and decision trees
5. Add/update patterns in catalog
6. Evaluate automation opportunities
7. Document learnings
8. Bump version if user-facing changes
9. Notify team of improvements

Improvement Triggers:

• Pattern appears 3+ times → Consider automation
• New anti-pattern discovered → Update guidance
• Workflow changes → Update playbooks
• Tool updates → Validate scripts

Cross-Agent Compatibility

For Claude Code Users

Invocation:

@skill {guru-id}
{Your request}

Triggers: Keywords like "{keyword1}", "{keyword2}", "{keyword3}"

For GitHub Copilot Users

Access:

• Read .agents/{guru-id}.md for equivalent guidance
• Run scripts directly: dotnet fsi .claude/skills/{guru-id}/scripts/{script}.fsx
• Follow decision trees and playbooks
• Reference pattern catalog

Quick Start:

# Run {common task}
dotnet fsi .claude/skills/{guru-id}/scripts/{script-name}.fsx

For Other Agents (Cursor, Windsurf, Aider)

Access:

• Documentation: .claude/skills/{guru-id}/skill.md (this file)
• Quick reference: .claude/skills/{guru-id}/README.md
• Scripts: .claude/skills/{guru-id}/scripts/
• Templates: .claude/skills/{guru-id}/templates/

Usage:

1. Read skill.md for comprehensive guidance
2. Follow playbooks for complex workflows
3. Use decision trees for problem-solving
4. Run automation scripts for common tasks
5. Reference pattern catalog for examples

Templates

Location: .claude/skills/{guru-id}/templates/

Template 1: {template-name-1}.md

Purpose: {What this template is for}
When to Use: {Scenarios for using this template}

Usage:

1. Copy template: cp .claude/skills/{guru-id}/templates/{template-name-1}.md .
2. Fill in {placeholder} values
3. Use for {intended purpose}

Template 2: {template-name-2}.md

{Similar structure to Template 1}

Related Resources

Within This Project:

• README.md - Quick reference guide
• MAINTENANCE.md - Maintenance and evolution guide
• metadata.yaml - Skill metadata and configuration
• scripts/ - Automation scripts
• templates/ - Reusable templates
• patterns/ - Detailed pattern documentation

Project Guidance:

• AGENTS.md - Primary agent guidance
• Guru Philosophy - Guru principles
• Skills Reference - All available gurus
• Skill Matrix - Maturity tracking

External Resources:

• {External resource 1} - {Description}
• {External resource 2} - {Description}
• {External resource 3} - {Description}

Last Updated: {YYYY-MM-DD}
Version: {Semantic version}
Status: {alpha/beta/stable}
Maintainer: {Maintainer name/GitHub handle}