claude-mpm 4.1.26__py3-none-any.whl → 4.24.0__py3-none-any.whl

claude_mpm/commands/mpm-agents-recommend.md

@@ -0,0 +1,214 @@
+# Show recommended agents for detected project stack
+
+Get intelligent agent recommendations based on your project's detected toolchain.
+
+## Usage
+
+```
+/mpm-agents-recommend
+```
+
+## Description
+
+This command analyzes your detected project stack and recommends the most appropriate agents, organized by priority:
+- **Essential agents**: Core agents required for your primary stack
+- **Recommended agents**: Complementary agents for full functionality
+- **Optional agents**: Specialized agents for detected tools
+
+Each recommendation includes an explanation of why that agent is suggested.
+
+## Implementation
+
+When you run `/mpm-agents-recommend`, the PM will execute:
+```bash
+claude-mpm agents recommend
+```
+
+This runs the detection phase and then applies recommendation rules to suggest the best agents for your specific technology stack.
+
+## Expected Output
+
+```
+🤖 Agent Recommendations
+=========================
+
+Based on detected stack:
+  Python 3.11, FastAPI 0.104.0, pytest, Docker, Vercel
+
+Essential Agents (Must Have):
+  ✓ fastapi-engineer
+    Reason: FastAPI framework detected - specialized agent for FastAPI development
+    Capabilities: FastAPI routes, Pydantic models, async endpoints, dependency injection
+
+  ✓ python-engineer
+    Reason: Python project - general Python development support
+    Capabilities: Python code, testing, debugging, package management
+
+  ✓ api-qa
+    Reason: API testing for FastAPI backend
+    Capabilities: API endpoint testing, validation, load testing, contract testing
+
+Recommended Agents (Strongly Suggested):
+  ○ docker-ops
+    Reason: Docker configuration detected
+    Capabilities: Docker builds, container management, docker-compose orchestration
+
+  ○ vercel-ops
+    Reason: Vercel deployment configuration found
+    Capabilities: Vercel deployments, serverless functions, domain management
+
+  ○ playwright-qa
+    Reason: Comprehensive E2E testing capability
+    Capabilities: Browser automation, visual testing, API testing
+
+Optional Agents (Nice to Have):
+  ○ local-ops-agent
+    Reason: Local development and PM2 process management
+    Capabilities: Local server management, port handling, PM2 operations
+
+  ○ security-agent
+    Reason: Security scanning and best practices
+    Capabilities: Dependency scanning, code security, auth patterns
+
+Summary:
+  Recommended: 3 essential + 2 recommended = 5 agents
+  Optional: 2 additional agents for enhanced capabilities
+
+  Total recommended deployment: 5 agents
+  Maximum useful deployment: 7 agents
+
+Next Steps:
+  1. Review recommendations above
+  2. Run '/mpm-auto-configure --preview' to see deployment plan
+  3. Run '/mpm-auto-configure' to deploy recommended agents
+  4. Or deploy individually: '/mpm-agents deploy <agent-name>'
+```
+
+## Recommendation Logic
+
+### Python Projects
+
+**If FastAPI detected:**
+- Essential: fastapi-engineer, python-engineer, api-qa
+- Recommended: docker-ops (if Docker), vercel-ops (if Vercel)
+
+**If Flask detected:**
+- Essential: flask-engineer, python-engineer, api-qa
+- Recommended: docker-ops (if Docker)
+
+**If Django detected:**
+- Essential: django-engineer, python-engineer, api-qa
+- Recommended: docker-ops (if Docker)
+
+**If pytest detected:**
+- Add to recommended: api-qa, playwright-qa
+
+### JavaScript/TypeScript Projects
+
+**If Next.js detected:**
+- Essential: nextjs-engineer, react-engineer, web-qa
+- Recommended: playwright-qa, vercel-ops (if Vercel)
+
+**If React detected (without Next.js):**
+- Essential: react-engineer, web-qa
+- Recommended: playwright-qa
+
+**If Vue detected:**
+- Essential: vue-engineer, web-qa
+- Recommended: playwright-qa
+
+**If Express detected:**
+- Essential: node-engineer, api-qa
+- Recommended: docker-ops (if Docker)
+
+**If Nest.js detected:**
+- Essential: nestjs-engineer, node-engineer, api-qa
+- Recommended: docker-ops (if Docker)
+
+### Full-Stack Projects
+
+**Python backend + React frontend:**
+- Essential: fastapi-engineer (or flask-engineer), python-engineer, react-engineer, api-qa, web-qa
+- Recommended: playwright-qa, docker-ops, local-ops-agent
+
+**Node.js backend + React frontend:**
+- Essential: node-engineer, react-engineer, api-qa, web-qa
+- Recommended: playwright-qa, docker-ops
+
+### Testing-Focused Projects
+
+**If Playwright detected:**
+- Add to recommended: playwright-qa
+
+**If Jest/Vitest detected:**
+- Ensure web-qa or api-qa in recommended
+
+### Deployment-Focused Projects
+
+**If Vercel detected:**
+- Add to recommended: vercel-ops
+
+**If Railway detected:**
+- Add to recommended: railway-ops
+
+**If Docker detected:**
+- Add to recommended: docker-ops
+
+**If PM2 detected:**
+- Add to recommended: local-ops-agent
+
+## Agent Descriptions
+
+### Backend Specialists
+- **fastapi-engineer**: FastAPI expert - routes, Pydantic, async, websockets
+- **flask-engineer**: Flask expert - blueprints, extensions, templates
+- **django-engineer**: Django expert - models, views, ORM, admin
+- **node-engineer**: Node.js expert - Express, async, streams
+- **nestjs-engineer**: NestJS expert - modules, decorators, DI
+
+### Frontend Specialists
+- **nextjs-engineer**: Next.js expert - app router, server components, SSR
+- **react-engineer**: React expert - hooks, state, components
+- **vue-engineer**: Vue expert - composition API, components
+
+### QA Specialists
+- **api-qa**: API testing expert - REST, GraphQL, validation
+- **web-qa**: Web testing expert - fetch, integration, E2E
+- **playwright-qa**: Browser automation expert - UI testing, visual regression
+
+### Ops Specialists
+- **docker-ops**: Docker expert - builds, compose, containers
+- **vercel-ops**: Vercel deployment expert - serverless, edge
+- **railway-ops**: Railway deployment expert
+- **local-ops-agent**: Local development expert - PM2, ports, processes
+
+## Use Cases
+
+1. **Planning**: Understand what agents you should deploy before starting work
+2. **Validation**: Verify that auto-configuration will suggest the right agents
+3. **Learning**: Discover what capabilities are available for your stack
+4. **Optimization**: Find additional agents that could enhance your workflow
+
+## Tips
+
+1. **Start here**: Run this before `/mpm-auto-configure` to preview recommendations
+2. **Essential vs Optional**: Focus on essential agents first, add others as needed
+3. **Stack-specific**: Recommendations are tailored to YOUR detected stack
+4. **Flexible**: You can always deploy additional agents later or skip some
+5. **Explanations**: Pay attention to the "Reason" for each recommendation
+
+## Customizing Recommendations
+
+After seeing recommendations, you can:
+1. **Accept all**: Run `/mpm-auto-configure` to deploy all recommended agents
+2. **Pick and choose**: Deploy specific agents with `/mpm-agents deploy <name>`
+3. **Skip optional**: Deploy only essential and recommended agents
+4. **Add more**: Deploy additional agents not in recommendations
+
+## Related Commands
+
+- `/mpm-agents-detect` - See what was detected in your project
+- `/mpm-auto-configure` - Automatically deploy recommended agents
+- `/mpm-agents deploy <name>` - Deploy specific agents manually
+- `/mpm-agents` - View all available and deployed agents
+- `/mpm-help agents` - Learn more about agent management

claude_mpm/commands/mpm-agents.md

@@ -2,11 +2,121 @@
 
 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
+## Usage
+
+```
+/mpm-agents
+```
+
+## Description
+
+This command lists all available Claude MPM agents, including both built-in agents and any custom agents you've created. It shows their current deployment status, version information, and capabilities.
+
+## What This Command Does
+
+When you run `/mpm-agents`, I will:
+
+1. **List Available Agents**: Run `claude-mpm agents list` to show all agents
+2. **Display Agent Information**:
+   - Agent names and IDs
+   - Brief descriptions
+   - Model preferences (opus, sonnet, haiku)
+   - Tool availability
+   - Version information
+   - Deployment status
+
+## Output Example
+
+The command displays agents in a formatted table showing:
+- Agent name and description
+- Version and model preference
+- Tools available to the agent
+- Current deployment status
+
+## Implementation
+
+To show available agents, I'll execute:
+```bash
+claude-mpm agents list --deployed
+```
+
+This will display all deployed agents that are currently available for use.
+
+Alternatively, you can use these variations:
+- `claude-mpm agents list --system` - Show system agents
+- `claude-mpm agents list --by-tier` - Group agents by precedence tier
+- `claude-mpm agents list --all` - Show all agents including undeployed
+
+## Auto-Configuration Subcommands (NEW!)
+
+### Quick Agent Setup
+
+Claude MPM now includes intelligent auto-configuration that detects your project and recommends the right agents:
+
+#### `agents detect`
+Scan your project to detect toolchain and frameworks:
+```bash
+claude-mpm agents detect
+```
+
+Shows detected:
+- Programming languages (Python, Node.js, Rust, Go, etc.)
+- Frameworks (FastAPI, Next.js, React, Express, etc.)
+- Testing tools (pytest, Jest, Playwright, etc.)
+- Build tools and package managers
+- Deployment configurations
+
+#### `agents recommend`
+Get agent recommendations based on detected toolchain:
+```bash
+claude-mpm agents recommend
+```
+
+Shows:
+- Essential agents for your stack
+- Recommended complementary agents
+- Optional specialized agents
+- Rationale for each recommendation
+
+#### `auto-configure`
+Automatically detect, recommend, and deploy agents:
+```bash
+claude-mpm auto-configure --preview  # See what would be configured
+claude-mpm auto-configure            # Interactive configuration
+claude-mpm auto-configure --yes      # Auto-apply without prompts
+```
+
+**Example workflow:**
+1. Run `claude-mpm agents detect` to see what's detected
+2. Run `claude-mpm agents recommend` to see suggestions
+3. Run `claude-mpm auto-configure` to apply configuration
+4. Or skip straight to `claude-mpm auto-configure --yes`
+
+### Supported Stacks
+
+**Python:**
+- FastAPI, Flask, Django → fastapi-engineer, python-engineer
+- pytest, unittest → api-qa, python-qa
+
+**JavaScript/TypeScript:**
+- Next.js → nextjs-engineer, react-engineer
+- React, Vue, Svelte → react-engineer, web-qa
+- Express, Nest.js → node-engineer, api-qa
+- Jest, Vitest, Playwright → playwright-qa, web-qa
+
+**Full-Stack Projects:**
+Automatically recommends both frontend and backend agents based on your complete stack.
+
+**Deployment:**
+- Vercel → vercel-ops
+- Railway → railway-ops
+- Docker → docker-ops
+- PM2 → local-ops-agent
+
+## Related Commands
+
+For more information on auto-configuration:
+- `/mpm-help auto-configure` - Detailed auto-configuration help
+- `/mpm-agents-detect` - Run detection via slash command
+- `/mpm-agents-recommend` - Show recommendations via slash command
+- `/mpm-auto-configure` - Run auto-configuration via slash command

claude_mpm/commands/mpm-auto-configure.md

@@ -0,0 +1,269 @@
+# Automatically configure agents based on project detection
+
+Automatically detect your project's toolchain and configure the most appropriate agents.
+
+## Usage
+
+```
+/mpm-auto-configure [options]
+```
+
+## Description
+
+This command provides intelligent auto-configuration that:
+1. Scans your project to detect programming languages, frameworks, and tools
+2. Recommends the most appropriate agents for your stack
+3. Optionally deploys the recommended agents with confirmation
+
+This is the fastest way to get started with Claude MPM in any project!
+
+## Options
+
+- `--preview` - Show what would be configured without making any changes
+- `--yes` - Automatically apply recommendations without prompting
+- `--force` - Force reconfiguration even if agents are already deployed
+
+## Implementation
+
+When you run `/mpm-auto-configure`, the PM will:
+
+1. **Detect Your Stack**:
+   - Scan for package.json, requirements.txt, Cargo.toml, go.mod, etc.
+   - Identify frameworks (FastAPI, Next.js, React, Express, etc.)
+   - Detect testing tools (pytest, Jest, Playwright, etc.)
+   - Find build tools and deployment configurations
+
+2. **Recommend Agents**:
+   - **Essential agents**: Required for your primary stack
+   - **Recommended agents**: Complementary agents for full functionality
+   - **Optional agents**: Specialized agents for detected tools
+
+3. **Deploy Agents** (with confirmation):
+   - Show what will be deployed
+   - Request confirmation (unless --yes is used)
+   - Deploy agents to your project
+   - Verify deployment success
+
+## Examples
+
+### Preview Mode (Recommended First Step)
+```
+/mpm-auto-configure --preview
+```
+Shows what would be configured without making changes. Great for understanding recommendations before applying.
+
+### Interactive Configuration
+```
+/mpm-auto-configure
+```
+Detect, recommend, and prompt for confirmation before deploying.
+
+### Automatic Configuration
+```
+/mpm-auto-configure --yes
+```
+Automatically apply all recommendations without prompting. Best for quick setup.
+
+### Force Reconfiguration
+```
+/mpm-auto-configure --force
+```
+Reconfigure agents even if some are already deployed. Useful for stack changes.
+
+## Expected Output
+
+```
+🤖 Auto-Configuration for Claude MPM
+=====================================
+
+Step 1: Detecting Project Stack
+--------------------------------
+✓ Detected Python 3.11
+✓ Detected FastAPI 0.104.0
+✓ Detected pytest 7.4.0
+✓ Detected Docker configuration
+✓ Detected Vercel deployment
+
+Step 2: Agent Recommendations
+------------------------------
+Essential Agents (3):
+  ✓ fastapi-engineer - FastAPI framework detected
+  ✓ python-engineer - Python project support
+  ✓ api-qa - API testing and validation
+
+Recommended Agents (2):
+  ○ docker-ops - Docker configuration found
+  ○ vercel-ops - Vercel deployment detected
+
+Optional Agents (1):
+  ○ playwright-qa - Browser testing capability
+
+Step 3: Deploy Agents
+---------------------
+Deploy 5 agents? (y/n): y
+
+Deploying agents...
+✓ fastapi-engineer deployed
+✓ python-engineer deployed
+✓ api-qa deployed
+✓ docker-ops deployed
+✓ vercel-ops deployed
+
+🎉 Auto-configuration complete!
+5 agents deployed successfully.
+
+Next steps:
+- Run /mpm-agents to see your deployed agents
+- Start working with specialized agents for your stack
+- Use /mpm-help for more information
+```
+
+## What Gets Detected
+
+### Languages
+- Python (CPython, PyPy)
+- JavaScript/TypeScript (Node.js, Deno, Bun)
+- Rust
+- Go
+- Java
+
+### Python Frameworks
+- FastAPI
+- Flask
+- Django
+- Starlette
+- Pyramid
+
+### JavaScript/TypeScript Frameworks
+- Next.js
+- React
+- Vue
+- Svelte
+- Angular
+- Express
+- Nest.js
+- Fastify
+
+### Testing Tools
+- pytest (Python)
+- unittest (Python)
+- Jest (JavaScript)
+- Vitest (JavaScript)
+- Playwright (Browser)
+- Cypress (Browser)
+
+### Build Tools
+- Vite
+- Webpack
+- Rollup
+- esbuild
+- Turbopack
+
+### Deployment Platforms
+- Vercel
+- Railway
+- Docker
+- PM2
+- Kubernetes
+
+## Agent Mapping Examples
+
+### Python + FastAPI
+**Essential:**
+- fastapi-engineer
+- python-engineer
+- api-qa
+
+**Recommended:**
+- docker-ops (if Docker detected)
+- vercel-ops (if Vercel detected)
+
+### Next.js + React
+**Essential:**
+- nextjs-engineer
+- react-engineer
+- web-qa
+
+**Recommended:**
+- playwright-qa (if Playwright detected)
+- vercel-ops (if Vercel detected)
+
+### Full-Stack (FastAPI + React)
+**Essential:**
+- fastapi-engineer
+- python-engineer
+- react-engineer
+- api-qa
+- web-qa
+
+**Recommended:**
+- playwright-qa
+- docker-ops
+- local-ops-agent
+
+## Default Configuration Fallback
+
+When auto-configuration cannot detect your project's toolchain (e.g., a new or uncommon language/framework), it automatically falls back to a sensible set of default general-purpose agents instead of leaving your project unconfigured.
+
+### When Defaults are Applied
+
+Default agents are deployed when:
+- Primary language is detected as "Unknown"
+- No specific framework or toolchain can be identified
+- No agent recommendations can be generated from detected technologies
+
+### Default Agents
+
+The following general-purpose agents are recommended with moderate confidence (0.7):
+
+- **engineer**: General-purpose engineer for code implementation
+- **research**: Code exploration and analysis
+- **qa**: Testing and quality assurance
+- **ops**: Infrastructure and deployment operations
+- **documentation**: Documentation and technical writing
+
+All default recommendations will be marked with `is_default: True` in the metadata.
+
+### Disabling Default Fallback
+
+To disable the default configuration fallback, edit `.claude-mpm/config/agent_capabilities.yaml`:
+
+```yaml
+default_configuration:
+  enabled: false  # Disable default fallback
+```
+
+When disabled, auto-configuration will recommend zero agents if the toolchain cannot be detected.
+
+### Customizing Defaults
+
+You can customize which agents are deployed by default by editing the `default_configuration.agents` section in `.claude-mpm/config/agent_capabilities.yaml`:
+
+```yaml
+default_configuration:
+  enabled: true
+  min_confidence: 0.7  # Confidence score for default recommendations
+  agents:
+    - agent_id: engineer
+      reasoning: "General-purpose engineer for code implementation"
+      priority: 1
+    - agent_id: research
+      reasoning: "Code exploration and analysis"
+      priority: 2
+    # Add or remove agents as needed
+```
+
+## Tips
+
+1. **Start with preview**: Always run with `--preview` first to see recommendations
+2. **Review carefully**: Check that detected stack matches your project
+3. **Customize later**: You can always deploy/remove agents manually after auto-config
+4. **Re-run after changes**: Run again with `--force` if you add new frameworks
+5. **Complementary commands**: Use `/mpm-agents-detect` and `/mpm-agents-recommend` for more details
+
+## Related Commands
+
+- `/mpm-agents-detect` - Just show detected toolchain
+- `/mpm-agents-recommend` - Show recommendations without deploying
+- `/mpm-agents` - Manage agents manually
+- `/mpm-help agents` - Learn about manual agent management