claude-mpm 4.1.14__py3-none-any.whl → 4.1.17__py3-none-any.whl

claude_mpm/VERSION

@@ -1 +1 @@
-4.1.14
+4.1.17

claude_mpm/agents/templates/ops.json

@@ -1,9 +1,14 @@
 {
   "schema_version": "1.2.0",
   "agent_id": "ops-agent",
-  "agent_version": "2.2.1",
-  "template_version": "2.1.0",
+  "agent_version": "2.2.2",
+  "template_version": "2.2.0",
   "template_changelog": [
+    {
+      "version": "2.2.0",
+      "date": "2025-08-29",
+      "description": "Added comprehensive git commit authority with mandatory security verification"
+    },
     {
       "version": "2.1.0",
       "date": "2025-08-25",
@@ -33,7 +38,7 @@
     ],
     "author": "Claude MPM Team",
     "created_at": "2025-07-27T03:45:51.476769Z",
-    "updated_at": "2025-08-22T12:00:00.000000Z",
+    "updated_at": "2025-08-29T12:00:00.000000Z",
     "color": "orange"
   },
   "capabilities": {
@@ -64,7 +69,7 @@
       ]
     }
   },
-  "instructions": "# Ops Agent\n\n**Inherits from**: BASE_AGENT_TEMPLATE.md\n**Focus**: Infrastructure automation and system operations\n\n## Core Expertise\n\nManage infrastructure, deployments, and system operations with a focus on reliability and automation. Handle CI/CD, monitoring, and operational excellence.\n\n## Ops-Specific Memory Management\n\n**Configuration Sampling**:\n- Extract patterns from config files, not full content\n- Use grep for environment variables and settings\n- Process deployment scripts sequentially\n- Sample 2-3 representative configs per service\n\n## Operations Protocol\n\n### Infrastructure Management\n```bash\n# Check system resources\ndf -h | head -10\nfree -h\nps aux | head -20\nnetstat -tlnp 2>/dev/null | head -10\n```\n\n### Deployment Operations\n```bash\n# Docker operations\ndocker ps --format \"table {{.Names}}\t{{.Status}}\t{{.Ports}}\"\ndocker images --format \"table {{.Repository}}\t{{.Tag}}\t{{.Size}}\"\n\n# Kubernetes operations (if applicable)\nkubectl get pods -o wide | head -20\nkubectl get services | head -10\n```\n\n### CI/CD Pipeline Management\n```bash\n# Check pipeline status\ngrep -r \"stage:\" .gitlab-ci.yml 2>/dev/null\ngrep -r \"jobs:\" .github/workflows/*.yml 2>/dev/null | head -10\n```\n\n## Operations Focus Areas\n\n- **Infrastructure**: Servers, containers, orchestration\n- **Deployment**: CI/CD pipelines, release management\n- **Monitoring**: Logs, metrics, alerts\n- **Security**: Access control, secrets management\n- **Performance**: Resource optimization, scaling\n- **Reliability**: Backup, recovery, high availability\n\n## Operations Categories\n\n### Infrastructure as Code\n- Terraform configurations\n- Ansible playbooks\n- CloudFormation templates\n- Kubernetes manifests\n\n### Monitoring & Observability\n- Log aggregation setup\n- Metrics collection\n- Alert configuration\n- Dashboard creation\n\n### Security Operations\n- Secret rotation\n- Access management\n- Security scanning\n- Compliance checks\n\n## Ops-Specific Todo Patterns\n\n**Infrastructure Tasks**:\n- `[Ops] Configure production deployment pipeline`\n- `[Ops] Set up monitoring for new service`\n- `[Ops] Implement auto-scaling rules`\n\n**Maintenance Tasks**:\n- `[Ops] Update SSL certificates`\n- `[Ops] Rotate database credentials`\n- `[Ops] Patch security vulnerabilities`\n\n**Optimization Tasks**:\n- `[Ops] Optimize container images`\n- `[Ops] Reduce infrastructure costs`\n- `[Ops] Improve deployment speed`\n\n## Operations Workflow\n\n### Phase 1: Assessment\n```bash\n# Check current state\ndocker-compose ps 2>/dev/null || docker ps\nsystemctl status nginx 2>/dev/null || service nginx status\ngrep -h \"ENV\" Dockerfile* 2>/dev/null | head -10\n```\n\n### Phase 2: Implementation\n```bash\n# Apply changes safely\n# Always backup before changes\n# Use --dry-run when available\n# Test in staging first\n```\n\n### Phase 3: Verification\n```bash\n# Verify deployments\ncurl -I http://localhost/health 2>/dev/null\ndocker logs app --tail=50 2>/dev/null\nkubectl rollout status deployment/app 2>/dev/null\n```\n\n## Ops Memory Categories\n\n**Pattern Memories**: Deployment patterns, config patterns\n**Architecture Memories**: Infrastructure topology, service mesh\n**Performance Memories**: Bottlenecks, optimization wins\n**Security Memories**: Vulnerabilities, security configs\n**Context Memories**: Environment specifics, tool versions\n\n## Operations Standards\n\n- **Automation**: Infrastructure as Code for everything\n- **Safety**: Always test in staging first\n- **Documentation**: Clear runbooks and procedures\n- **Monitoring**: Comprehensive observability\n- **Security**: Defense in depth approach",
+  "instructions": "# Ops Agent\n\n**Inherits from**: BASE_AGENT_TEMPLATE.md\n**Focus**: Infrastructure automation and system operations\n\n## Core Expertise\n\nManage infrastructure, deployments, and system operations with a focus on reliability and automation. Handle CI/CD, monitoring, and operational excellence.\n\n## Ops-Specific Memory Management\n\n**Configuration Sampling**:\n- Extract patterns from config files, not full content\n- Use grep for environment variables and settings\n- Process deployment scripts sequentially\n- Sample 2-3 representative configs per service\n\n## Operations Protocol\n\n### Infrastructure Management\n```bash\n# Check system resources\ndf -h | head -10\nfree -h\nps aux | head -20\nnetstat -tlnp 2>/dev/null | head -10\n```\n\n### Deployment Operations\n```bash\n# Docker operations\ndocker ps --format \"table {{.Names}}\t{{.Status}}\t{{.Ports}}\"\ndocker images --format \"table {{.Repository}}\t{{.Tag}}\t{{.Size}}\"\n\n# Kubernetes operations (if applicable)\nkubectl get pods -o wide | head -20\nkubectl get services | head -10\n```\n\n### CI/CD Pipeline Management\n```bash\n# Check pipeline status\ngrep -r \"stage:\" .gitlab-ci.yml 2>/dev/null\ngrep -r \"jobs:\" .github/workflows/*.yml 2>/dev/null | head -10\n```\n\n## Operations Focus Areas\n\n- **Infrastructure**: Servers, containers, orchestration\n- **Deployment**: CI/CD pipelines, release management\n- **Monitoring**: Logs, metrics, alerts\n- **Security**: Access control, secrets management\n- **Performance**: Resource optimization, scaling\n- **Reliability**: Backup, recovery, high availability\n\n## Operations Categories\n\n### Infrastructure as Code\n- Terraform configurations\n- Ansible playbooks\n- CloudFormation templates\n- Kubernetes manifests\n\n### Monitoring & Observability\n- Log aggregation setup\n- Metrics collection\n- Alert configuration\n- Dashboard creation\n\n### Security Operations\n- Secret rotation\n- Access management\n- Security scanning\n- Compliance checks\n\n## Ops-Specific Todo Patterns\n\n**Infrastructure Tasks**:\n- `[Ops] Configure production deployment pipeline`\n- `[Ops] Set up monitoring for new service`\n- `[Ops] Implement auto-scaling rules`\n\n**Maintenance Tasks**:\n- `[Ops] Update SSL certificates`\n- `[Ops] Rotate database credentials`\n- `[Ops] Patch security vulnerabilities`\n\n**Optimization Tasks**:\n- `[Ops] Optimize container images`\n- `[Ops] Reduce infrastructure costs`\n- `[Ops] Improve deployment speed`\n\n## Operations Workflow\n\n### Phase 1: Assessment\n```bash\n# Check current state\ndocker-compose ps 2>/dev/null || docker ps\nsystemctl status nginx 2>/dev/null || service nginx status\ngrep -h \"ENV\" Dockerfile* 2>/dev/null | head -10\n```\n\n### Phase 2: Implementation\n```bash\n# Apply changes safely\n# Always backup before changes\n# Use --dry-run when available\n# Test in staging first\n```\n\n### Phase 3: Verification\n```bash\n# Verify deployments\ncurl -I http://localhost/health 2>/dev/null\ndocker logs app --tail=50 2>/dev/null\nkubectl rollout status deployment/app 2>/dev/null\n```\n\n## Ops Memory Categories\n\n**Pattern Memories**: Deployment patterns, config patterns\n**Architecture Memories**: Infrastructure topology, service mesh\n**Performance Memories**: Bottlenecks, optimization wins\n**Security Memories**: Vulnerabilities, security configs\n**Context Memories**: Environment specifics, tool versions\n\n## Git Commit Authority\n\nThe Ops agent has full authority to make git commits for infrastructure, deployment, and operational changes with mandatory security verification.\n\n### Pre-Commit Security Protocol\n\n**MANDATORY**: Before ANY git commit, you MUST:\n1. Run security scans to detect secrets/keys\n2. Verify no sensitive data in staged files\n3. Check for hardcoded credentials\n4. Ensure environment variables are externalized\n\n### Security Verification Commands\n\nAlways run these checks before committing:\n```bash\n# 1. Use existing security infrastructure\nmake quality  # Runs bandit and other security checks\n\n# 2. Additional secret pattern detection\n# Check for API keys and tokens\nrg -i \"(api[_-]?key|token|secret|password)\\s*[=:]\\s*['\\\"][^'\\\"]{10,}\" --type-add 'config:*.{json,yaml,yml,toml,ini,env}' -tconfig -tpy\n\n# Check for AWS keys\nrg \"AKIA[0-9A-Z]{16}\" .\n\n# Check for private keys\nrg \"-----BEGIN (RSA |EC |OPENSSH |DSA |)?(PRIVATE|SECRET) KEY-----\" .\n\n# Check for high-entropy strings (potential secrets)\nrg \"['\\\"][A-Za-z0-9+/]{40,}[=]{0,2}['\\\"]\" --type-add 'config:*.{json,yaml,yml,toml,ini}' -tconfig\n\n# 3. Verify no large binary files\nfind . -type f -size +1000k -not -path \"./.git/*\" -not -path \"./node_modules/*\"\n```\n\n### Git Commit Workflow\n\n1. **Stage Changes**:\n   ```bash\n   git add <specific-files>  # Prefer specific files over git add .\n   ```\n\n2. **Security Verification**:\n   ```bash\n   # Run full security scan\n   make quality\n   \n   # If make quality not available, run manual checks\n   git diff --cached --name-only | xargs -I {} sh -c 'echo \"Checking {}\" && rg -i \"password|secret|token|api.key\" {} || true'\n   ```\n\n3. **Commit with Structured Message**:\n   ```bash\n   git commit -m \"type(scope): description\n   \n   - Detail 1\n   - Detail 2\n   \n   🤖 Generated with [Claude Code](https://claude.ai/code)\n   \n   Co-Authored-By: Claude <noreply@anthropic.com>\"\n   ```\n\n### Prohibited Patterns\n\n**NEVER commit files containing**:\n- Hardcoded passwords: `password = \"actual_password\"`\n- API keys: `api_key = \"sk-...\"`\n- Private keys: `-----BEGIN PRIVATE KEY-----`\n- Database URLs with credentials: `postgresql://user:pass@host`\n- AWS/Cloud credentials: `AKIA...` patterns\n- JWT tokens: `eyJ...` patterns\n- .env files with actual values (use .env.example instead)\n\n### Security Response Protocol\n\nIf secrets are detected:\n1. **STOP** - Do not proceed with commit\n2. **Remove** - Clean the sensitive data\n3. **Externalize** - Move to environment variables\n4. **Document** - Update .env.example with placeholders\n5. **Verify** - Re-run security checks\n6. **Commit** - Only after all checks pass\n\n### Commit Types (Conventional Commits)\n\nUse these prefixes for infrastructure commits:\n- `feat:` New infrastructure features\n- `fix:` Infrastructure bug fixes\n- `perf:` Performance improvements\n- `refactor:` Infrastructure refactoring\n- `docs:` Documentation updates\n- `chore:` Maintenance tasks\n- `ci:` CI/CD pipeline changes\n- `build:` Build system changes\n- `revert:` Revert previous commits\n\n## Operations Standards\n\n- **Automation**: Infrastructure as Code for everything\n- **Safety**: Always test in staging first\n- **Documentation**: Clear runbooks and procedures\n- **Monitoring**: Comprehensive observability\n- **Security**: Defense in depth approach",
   "knowledge": {
     "domain_expertise": [
       "Docker and container orchestration",
@@ -131,7 +136,9 @@
       "Deployment patterns and rollback procedures",
       "Infrastructure configurations",
       "Monitoring and alerting strategies",
-      "CI/CD pipeline requirements"
+      "CI/CD pipeline requirements",
+      "Git commit protocols and security verification",
+      "Security scanning patterns and secret detection"
     ],
     "keywords": [
       "deployment",
@@ -150,7 +157,9 @@
       "grafana",
       "aws",
       "azure",
-      "gcp"
+      "gcp",
+      "git commit protocols",
+      "security scanning patterns"
     ]
   },
   "dependencies": {

claude_mpm/cli/commands/mpm_init.py

@@ -3,6 +3,9 @@ MPM-Init Command - Initialize projects for optimal Claude Code and Claude MPM su
 
 This command delegates to the Agentic Coder Optimizer agent to establish clear,
 single-path project standards for documentation, tooling, and workflows.
+
+Enhanced with AST inspection capabilities for generating comprehensive developer
+documentation with code structure analysis.
 """
 
 import logging
@@ -35,6 +38,7 @@ class MPMInitCommand:
         force: bool = False,
         verbose: bool = False,
         use_venv: bool = False,
+        ast_analysis: bool = True,
     ) -> Dict:
         """
         Initialize project with Agentic Coder Optimizer standards.
@@ -44,6 +48,8 @@ class MPMInitCommand:
             framework: Specific framework if applicable
             force: Force initialization even if project already configured
             verbose: Show detailed output
+            use_venv: Force use of venv instead of mamba
+            ast_analysis: Enable AST analysis for enhanced documentation
 
         Returns:
             Dict containing initialization results
@@ -59,7 +65,9 @@ class MPMInitCommand:
                 return {"status": "cancelled", "message": "Initialization cancelled"}
 
             # Build the delegation prompt
-            prompt = self._build_initialization_prompt(project_type, framework)
+            prompt = self._build_initialization_prompt(
+                project_type, framework, ast_analysis
+            )
 
             # Show initialization plan
             console.print(
@@ -71,8 +79,15 @@ class MPMInitCommand:
                     "• Optimized project structure\n"
                     "• Tool configurations (linting, formatting, testing)\n"
                     "• GitHub workflows and CI/CD setup\n"
-                    "• Memory system initialization\n\n"
-                    "[dim]Powered by Agentic Coder Optimizer Agent[/dim]",
+                    "• Memory system initialization\n"
+                    + (
+                        "• AST analysis for comprehensive code documentation\n"
+                        if ast_analysis
+                        else ""
+                    )
+                    + "• Holistic CLAUDE.md organization with ranked instructions\n"
+                    + "• Priority-based content structure (🔴🟡🟢⚪)\n"
+                    + "\n[dim]Powered by Agentic Coder Optimizer Agent[/dim]",
                     title="MPM-Init",
                     border_style="cyan",
                 )
@@ -111,7 +126,10 @@ class MPMInitCommand:
         return Path("claude-mpm")
 
     def _build_initialization_prompt(
-        self, project_type: Optional[str] = None, framework: Optional[str] = None
+        self,
+        project_type: Optional[str] = None,
+        framework: Optional[str] = None,
+        ast_analysis: bool = True,
     ) -> str:
         """Build the initialization prompt for the agent."""
         base_prompt = f"""Please delegate this task to the Agentic Coder Optimizer agent:
@@ -173,9 +191,122 @@ Please perform the following initialization tasks:
    - Step-by-step setup instructions
    - Common commands reference
    - Troubleshooting guide
+"""
+
+        if ast_analysis:
+            base_prompt += """
+9. **Perform AST Analysis** (using Code Analyzer agent if needed):
+   - Parse code files to extract structure (classes, functions, methods)
+   - Generate comprehensive API documentation
+   - Create code architecture diagrams
+   - Document function signatures and dependencies
+   - Extract docstrings and inline comments
+   - Map code relationships and inheritance hierarchies
+   - Generate developer documentation with:
+     * Module overview and purpose
+     * Class hierarchies and relationships
+     * Function/method documentation
+     * Type annotations and parameter descriptions
+     * Code complexity metrics
+     * Dependency graphs
+   - Create DEVELOPER.md with technical architecture details
+   - Add CODE_STRUCTURE.md with AST-derived insights
+"""
+
+        base_prompt += """
+
+10. **Holistic CLAUDE.md Organization** (CRITICAL - Do this LAST):
+   After completing all initialization tasks, take a holistic look at the CLAUDE.md file and:
+   
+   a) **Reorganize Content by Priority**:
+      - CRITICAL instructions (security, data handling, core business rules) at the TOP
+      - Project overview and purpose
+      - Key architectural decisions and constraints
+      - Development guidelines and standards
+      - Common tasks and workflows
+      - Links to additional documentation
+      - Nice-to-have or optional information at the BOTTOM
+   
+   b) **Rank Instructions by Importance**:
+      - Use clear markers: 
+        * 🔴 CRITICAL: Security, data handling, breaking changes, core business rules
+        * 🟡 IMPORTANT: Key workflows, architecture decisions, performance requirements
+        * 🟢 STANDARD: Common operations, coding standards, best practices
+        * ⚪ OPTIONAL: Nice-to-have features, experimental code, future considerations
+      - Group related instructions together
+      - Ensure no contradictory instructions exist
+      - Remove redundant or outdated information
+      - Add a "Priority Index" at the top listing all CRITICAL and IMPORTANT items
+   
+   c) **Optimize for AI Agent Understanding**:
+      - Use consistent formatting and structure
+      - Provide clear examples for complex instructions
+      - Include "WHY" explanations for critical rules
+      - Add quick reference sections for common operations
+      - Ensure instructions are actionable and unambiguous
+   
+   d) **Validate Completeness**:
+      - Ensure ALL critical project knowledge is captured
+      - Verify single-path principle (ONE way to do each task)
+      - Check that all referenced documentation exists
+      - Confirm all tools and dependencies are documented
+      - Test that a new AI agent could understand the project from CLAUDE.md alone
+   
+   e) **Add Meta-Instructions Section**:
+      - Include a section about how to maintain CLAUDE.md
+      - Document when and how to update instructions
+      - Provide guidelines for instruction priority levels
+      - Add a changelog or last-updated timestamp
+   
+   f) **Follow This CLAUDE.md Template Structure**:
+      ```markdown
+      # Project Name - CLAUDE.md
+      
+      ## 🎯 Priority Index
+      ### 🔴 CRITICAL Instructions
+      - [List all critical items with links to their sections]
+      
+      ### 🟡 IMPORTANT Instructions
+      - [List all important items with links to their sections]
+      
+      ## 📋 Project Overview
+      [Brief description and purpose]
+      
+      ## 🔴 CRITICAL: Security & Data Handling
+      [Critical security rules and data handling requirements]
+      
+      ## 🔴 CRITICAL: Core Business Rules
+      [Non-negotiable business logic and constraints]
+      
+      ## 🟡 IMPORTANT: Architecture & Design
+      [Key architectural decisions and patterns]
+      
+      ## 🟡 IMPORTANT: Development Workflow
+      ### ONE Way to Build
+      ### ONE Way to Test
+      ### ONE Way to Deploy
+      
+      ## 🟢 STANDARD: Coding Guidelines
+      [Standard practices and conventions]
+      
+      ## 🟢 STANDARD: Common Tasks
+      [How to perform routine operations]
+      
+      ## 📚 Documentation Links
+      [Links to additional resources]
+      
+      ## ⚪ OPTIONAL: Future Enhancements
+      [Nice-to-have features and ideas]
+      
+      ## 📝 Meta: Maintaining This Document
+      - Last Updated: [timestamp]
+      - Update Frequency: [when to update]
+      - Priority Guidelines: [how to assign priorities]
+      ```
 
 Please ensure all documentation is clear, concise, and optimized for AI agents to understand and follow.
 Focus on establishing ONE clear way to do ANYTHING in the project.
+The final CLAUDE.md should be a comprehensive, well-organized guide that any AI agent can follow to work effectively on this project.
 """
 
         return base_prompt
@@ -369,7 +500,11 @@ Focus on establishing ONE clear way to do ANYTHING in the project.
                     "[green]Your project is now optimized for Claude Code and Claude MPM![/green]\n\n"
                     "Key files:\n"
                     "• [cyan]CLAUDE.md[/cyan] - Main documentation for AI agents\n"
-                    "• [cyan].claude-mpm/[/cyan] - Configuration and memories\n\n"
+                    "  - Organized with priority rankings (🔴🟡🟢⚪)\n"
+                    "  - Instructions ranked by importance for AI understanding\n"
+                    "  - Holistic documentation review completed\n"
+                    "• [cyan].claude-mpm/[/cyan] - Configuration and memories\n"
+                    "• [cyan]CODE_STRUCTURE.md[/cyan] - AST-derived architecture documentation (if enabled)\n\n"
                     "[dim]Run 'claude-mpm run' to start using the optimized setup[/dim]",
                     title="Success",
                     border_style="green",
@@ -398,13 +533,18 @@ Focus on establishing ONE clear way to do ANYTHING in the project.
 @click.option(
     "--verbose", is_flag=True, help="Show detailed output during initialization"
 )
+@click.option(
+    "--ast-analysis/--no-ast-analysis",
+    default=True,
+    help="Enable/disable AST analysis for enhanced documentation (default: enabled)",
+)
 @click.argument(
     "project_path",
     type=click.Path(exists=True, file_okay=False, dir_okay=True),
     required=False,
     default=".",
 )
-def mpm_init(project_type, framework, force, verbose, project_path):
+def mpm_init(project_type, framework, force, verbose, ast_analysis, project_path):
     """
     Initialize a project for optimal use with Claude Code and Claude MPM.
 
@@ -414,11 +554,14 @@ def mpm_init(project_type, framework, force, verbose, project_path):
     - Configure development tools and standards
     - Set up memory systems for project knowledge
     - Optimize for AI agent understanding
+    - Perform AST analysis for enhanced developer documentation
 
     Examples:
         claude-mpm mpm-init
         claude-mpm mpm-init --project-type web --framework react
         claude-mpm mpm-init /path/to/project --force
+        claude-mpm mpm-init --ast-analysis  # Enable AST analysis (default)
+        claude-mpm mpm-init --no-ast-analysis  # Disable AST analysis
     """
     try:
         # Create command instance
@@ -426,7 +569,11 @@ def mpm_init(project_type, framework, force, verbose, project_path):
 
         # Run initialization (now synchronous)
         result = command.initialize_project(
-            project_type=project_type, framework=framework, force=force, verbose=verbose
+            project_type=project_type,
+            framework=framework,
+            force=force,
+            verbose=verbose,
+            ast_analysis=ast_analysis,
         )
 
         # Exit with appropriate code

claude_mpm/cli/commands/mpm_init_handler.py

@@ -56,6 +56,7 @@ def manage_mpm_init(args):
             "force": getattr(args, "force", False),
             "verbose": getattr(args, "verbose", False),
             "use_venv": getattr(args, "use_venv", False),
+            "ast_analysis": getattr(args, "ast_analysis", True),
         }
 
         # Execute initialization (now synchronous)

claude_mpm/cli/parsers/mpm_init_parser.py

@@ -88,6 +88,19 @@ def add_mpm_init_subparser(subparsers: Any) -> None:
         action="store_true",
         help="Use traditional Python venv instead of mamba/conda environment",
     )
+    init_group.add_argument(
+        "--ast-analysis",
+        action="store_true",
+        default=True,
+        dest="ast_analysis",
+        help="Enable AST analysis for enhanced developer documentation (default: enabled)",
+    )
+    init_group.add_argument(
+        "--no-ast-analysis",
+        action="store_false",
+        dest="ast_analysis",
+        help="Disable AST analysis for documentation generation",
+    )
 
     # Template options
     template_group = mpm_init_parser.add_argument_group("template options")

claude_mpm/commands/__init__.py

@@ -0,0 +1,14 @@
+"""
+Commands package for Claude MPM.
+
+This module contains command documentation files (Markdown) that are accessed
+via importlib.resources.path() for command help and reference material.
+
+This __init__.py file is required to make this directory recognizable as a
+Python package by setuptools, which is necessary for proper distribution
+and importlib resource access in installed packages.
+
+Without this file, importlib.resources.path("claude_mpm", "commands") would
+fail in the distributed package, even though it works in development when
+the package is installed in editable mode.
+"""

claude_mpm/commands/mpm-agents.md

@@ -0,0 +1,12 @@
+# Show available agents and their versions
+
+Show all available Claude MPM agents with their versions and deployment status.
+
+This command displays:
+- Agent names and descriptions
+- Version information
+- Tool availability
+- Model preferences
+- Deployment status
+
+Usage: /mpm-agents

claude_mpm/commands/mpm-config.md

@@ -0,0 +1,18 @@
+# View or modify claude-mpm configuration
+
+Manage Claude MPM configuration settings.
+
+Usage: /mpm-config [key] [value]
+
+Examples:
+- /mpm-config - Show all configuration
+- /mpm-config list - List all configuration keys
+- /mpm-config set logging debug - Set logging level to debug
+- /mpm-config get websocket-port - Get WebSocket port setting
+
+Configuration categories:
+- logging: Logging levels and output
+- websocket: WebSocket server settings
+- hooks: Hook service configuration
+- memory: Memory system settings
+- agents: Agent deployment options

claude_mpm/commands/mpm-doctor.md

@@ -0,0 +1,24 @@
+# Run diagnostic checks on claude-mpm installation
+
+Run comprehensive diagnostic checks on your Claude MPM installation to identify and fix common issues.
+
+This command checks:
+- Installation integrity
+- Configuration validity
+- WebSocket connectivity
+- Agent deployment status
+- Memory system health
+- Hook service status
+
+Usage: /mpm-doctor [options]
+
+Options:
+- --verbose: Show detailed output
+- --no-color: Disable colored output  
+- --checks [list]: Run specific checks only
+- --fix: Attempt to fix identified issues
+
+Examples:
+- /mpm-doctor
+- /mpm-doctor --verbose
+- /mpm-doctor --fix

claude_mpm/commands/mpm-help.md

@@ -0,0 +1,10 @@
+# Show help for available MPM commands
+
+Display help information for Claude MPM commands.
+
+Usage: /mpm-help [command]
+
+Examples:
+- /mpm-help - Show all MPM commands
+- /mpm-help doctor - Show detailed help for doctor command
+- /mpm-help agents - Show detailed help for agents command

claude_mpm/commands/mpm-init.md

@@ -0,0 +1,162 @@
+# /mpm-init
+
+Initialize your project for optimal use with Claude Code and Claude MPM using the Agentic Coder Optimizer agent.
+
+## Usage
+
+```
+/mpm-init
+/mpm-init --project-type web --framework react
+/mpm-init --force
+/mpm-init --ast-analysis
+/mpm-init --comprehensive
+```
+
+## Description
+
+This command delegates to the Agentic Coder Optimizer agent to establish clear, single-path project standards for documentation, tooling, and workflows. It creates comprehensive documentation optimized for AI agents.
+
+## Features
+
+- **📚 Comprehensive CLAUDE.md**: Creates AI-optimized project documentation
+- **🎯 Priority-based Organization**: Ranks instructions by importance (🔴🟡🟢⚪)
+- **🔍 AST Analysis**: Deep code structure analysis for enhanced documentation
+- **🚀 Single-path Workflows**: Establishes ONE way to do ANYTHING
+- **🧠 Memory System**: Initializes project knowledge retention
+- **🔧 Tool Configuration**: Sets up linting, formatting, testing
+- **📝 Holistic Review**: Final organization and validation pass
+
+## Options
+
+- `--project-type [type]`: Specify project type (web, api, cli, library, etc.)
+- `--framework [name]`: Specify framework (react, vue, django, fastapi, etc.)
+- `--force`: Force reinitialization even if project is already configured
+- `--ast-analysis`: Enable AST analysis for enhanced documentation (default: enabled)
+- `--no-ast-analysis`: Disable AST analysis for faster initialization
+- `--comprehensive`: Create comprehensive setup including CI/CD and deployment
+- `--minimal`: Create minimal configuration (CLAUDE.md only)
+
+## What This Command Does
+
+### 1. Project Analysis
+- Scans project structure and existing configurations
+- Identifies project type, language, and frameworks
+- Checks for existing documentation and tooling
+
+### 2. CLAUDE.md Creation/Update
+The command creates a well-organized CLAUDE.md with:
+
+```markdown
+## 🎯 Priority Index
+### 🔴 CRITICAL Instructions
+- Security rules, data handling, core business logic
+
+### 🟡 IMPORTANT Instructions  
+- Key workflows, architecture decisions
+
+### 🟢 STANDARD Instructions
+- Common operations, coding standards
+
+### ⚪ OPTIONAL Instructions
+- Nice-to-have features, future enhancements
+```
+
+### 3. Single-Path Standards
+- ONE command for building: `make build`
+- ONE command for testing: `make test`
+- ONE command for deployment: `make deploy`
+- Clear documentation of THE way to do things
+
+### 4. AST Analysis (Optional)
+When enabled, performs:
+- Code structure extraction (classes, functions, methods)
+- API documentation generation
+- Architecture diagram creation
+- Function signature and dependency mapping
+- Creates DEVELOPER.md with technical details
+- Adds CODE_STRUCTURE.md with AST insights
+
+### 5. Tool Configuration
+- Linting setup and configuration
+- Code formatting standards
+- Testing framework setup
+- Pre-commit hooks if needed
+
+### 6. Memory System
+- Creates `.claude-mpm/memories/` directory
+- Initializes memory files for project knowledge
+- Documents memory usage patterns
+
+### 7. Holistic Organization (Final Step)
+After all tasks, performs a comprehensive review:
+- Reorganizes content by priority
+- Validates completeness
+- Ensures single-path principle
+- Adds meta-instructions for maintenance
+
+## Examples
+
+### Basic Initialization
+```bash
+/mpm-init
+```
+Analyzes current directory and creates optimal setup.
+
+### Web Project with React
+```bash
+/mpm-init --project-type web --framework react
+```
+Initializes with web-specific configurations and React patterns.
+
+### Force Reinitialization
+```bash
+/mpm-init --force --comprehensive
+```
+Overwrites existing configuration with comprehensive setup.
+
+### Fast Mode (No AST)
+```bash
+/mpm-init --no-ast-analysis --minimal
+```
+Quick initialization without code analysis.
+
+## Implementation
+
+This command executes:
+```bash
+claude-mpm mpm-init [options]
+```
+
+The command delegates to the Agentic Coder Optimizer agent which:
+1. Analyzes your project structure
+2. Creates comprehensive documentation
+3. Establishes single-path workflows
+4. Configures development tools
+5. Sets up memory systems
+6. Performs AST analysis (if enabled)
+7. Organizes everything with priority rankings
+
+## Expected Output
+
+After successful execution:
+- ✅ **CLAUDE.md**: Main AI agent documentation with priority rankings
+- ✅ **Single-path workflows**: Clear commands for all operations
+- ✅ **Tool configurations**: Linting, formatting, testing setup
+- ✅ **Memory system**: Initialized for knowledge retention
+- ✅ **Developer docs**: Technical documentation (with AST analysis)
+- ✅ **Priority organization**: Instructions ranked by importance
+
+## Notes
+
+- The command uses the Agentic Coder Optimizer agent for implementation
+- AST analysis is enabled by default for comprehensive documentation
+- Priority rankings help AI agents focus on critical instructions first
+- The holistic review ensures documentation quality and completeness
+- All documentation is optimized for AI agent understanding
+
+## Related Commands
+
+- `/mpm-status`: Check current project setup status
+- `/mpm-agents`: Manage specialized agents
+- `/mpm-config`: Configure Claude MPM settings
+- `/mpm-doctor`: Diagnose and fix issues

claude_mpm/commands/mpm-status.md

@@ -0,0 +1,13 @@
+# Show claude-mpm status and environment
+
+Display the current status of Claude MPM including environment information, active services, and system health.
+
+This command shows:
+- Claude MPM version
+- Python environment
+- Active services (WebSocket, Hook Service, etc.)
+- Memory usage
+- Agent deployment status
+- Configuration summary
+
+Usage: /mpm-status