claude-mpm 4.16.0__py3-none-any.whl → 4.20.3__py3-none-any.whl

claude_mpm/cli/commands/mpm_init.py

@@ -1952,19 +1952,24 @@ def context_command(session_id, days, project_path):
         sys.exit(1)
 
 
-# Add deprecated 'resume' alias for backward compatibility
-@mpm_init.command(name="resume", hidden=True)
+# Resume command - NEW: reads from stop event logs
+@mpm_init.command(name="resume")
+@click.option(
+    "--list",
+    "list_sessions",
+    is_flag=True,
+    help="List available sessions from logs",
+)
 @click.option(
     "--session-id",
-    "-i",
+    "-s",
     type=str,
-    help="Unused (for compatibility) - will be removed in future version",
+    help="Resume specific session by ID",
 )
 @click.option(
-    "--days",
+    "--last",
     type=int,
-    default=7,
-    help="Number of days of git history to analyze (default: 7)",
+    help="Show last N sessions",
 )
 @click.argument(
     "project_path",
@@ -1972,35 +1977,115 @@ def context_command(session_id, days, project_path):
     required=False,
     default=".",
 )
-def resume_session(session_id, days, project_path):
+def resume_command(list_sessions, session_id, last, project_path):
     """
-    [DEPRECATED] Use 'context' instead.
+    Resume work from previous session using stop event logs.
+
+    Reads from:
+    - .claude-mpm/resume-logs/ (structured summaries, preferred)
+    - .claude-mpm/responses/ (raw conversation logs, fallback)
 
-    This command is deprecated and will be removed in a future version.
-    Please use 'claude-mpm mpm-init context' instead.
+    Examples:
+        claude-mpm mpm-init resume                    # Show latest session
+        claude-mpm mpm-init resume --list             # List all sessions
+        claude-mpm mpm-init resume --session-id ID    # Resume specific session
+        claude-mpm mpm-init resume --last 5           # Show last 5 sessions
     """
-    console.print(
-        "[yellow]⚠️  Warning: 'resume' is deprecated. Use 'context' instead.[/yellow]"
-    )
-    console.print("[dim]Run: claude-mpm mpm-init context[/dim]\n")
+    from claude_mpm.services.cli.resume_service import ResumeService
 
     try:
-        command = MPMInitCommand(Path(project_path))
-        result = command.handle_context(session_id=session_id, days=days)
+        service = ResumeService(Path(project_path))
 
-        if (
-            result["status"] == OperationResult.SUCCESS
-            or result["status"] == OperationResult.CONTEXT_READY
-        ):
+        # Handle --list flag
+        if list_sessions:
+            sessions = service.list_sessions()
+            if not sessions:
+                console.print("[yellow]No sessions found in response logs.[/yellow]")
+                console.print(
+                    "[dim]Sessions are stored in .claude-mpm/responses/[/dim]\n"
+                )
+                sys.exit(1)
+
+            # Limit by --last if specified
+            if last and last > 0:
+                sessions = sessions[:last]
+
+            console.print(
+                f"\n[bold cyan]📋 Available Sessions ({len(sessions)})[/bold cyan]\n"
+            )
+
+            from rich.table import Table
+
+            table = Table(show_header=True, header_style="bold magenta")
+            table.add_column("Session ID", style="cyan", width=25)
+            table.add_column("Time", style="yellow", width=20)
+            table.add_column("Agent", style="green", width=15)
+            table.add_column("Stop Reason", style="white", width=20)
+            table.add_column("Tokens", style="dim", width=10)
+
+            for session in sessions:
+                time_str = session.timestamp.strftime("%Y-%m-%d %H:%M")
+                tokens_str = (
+                    f"{session.token_usage // 1000}k"
+                    if session.token_usage > 0
+                    else "-"
+                )
+
+                table.add_row(
+                    session.session_id,
+                    time_str,
+                    session.last_agent,
+                    session.stop_reason,
+                    tokens_str,
+                )
+
+            console.print(table)
+            console.print()
             sys.exit(0)
+
+        # Handle --session-id
+        if session_id:
+            context = service.get_session_context(session_id)
+            if not context:
+                console.print(f"[red]Session '{session_id}' not found.[/red]")
+                console.print("[dim]Use --list to see available sessions.[/dim]\n")
+                sys.exit(1)
         else:
-            sys.exit(1)
+            # Default: get latest session
+            context = service.get_latest_session()
+            if not context:
+                console.print("[yellow]No sessions found in logs.[/yellow]")
+                console.print(
+                    "[dim]Sessions are stored in .claude-mpm/responses/[/dim]\n"
+                )
+                sys.exit(1)
+
+        # Display context
+        display_text = service.format_resume_display(context)
+        console.print(display_text)
+
+        # Ask if user wants to continue
+        from rich.prompt import Confirm
+
+        should_continue = Confirm.ask(
+            "\n[bold]Would you like to continue this work?[/bold]", default=True
+        )
+
+        if should_continue:
+            console.print(
+                "\n[green]✅ Great! Use this context to continue your work.[/green]\n"
+            )
+            sys.exit(0)
+        else:
+            console.print("\n[cyan]Starting fresh session instead.[/cyan]\n")
+            sys.exit(0)
 
     except KeyboardInterrupt:
-        console.print("\n[yellow]Context analysis cancelled by user[/yellow]")
+        console.print("\n[yellow]Resume cancelled by user[/yellow]")
         sys.exit(130)
     except Exception as e:
-        console.print(f"[red]Context analysis failed: {e}[/red]")
+        logger.error(f"Resume failed: {e}")
+        console.print(f"[red]Resume failed: {e}[/red]")
         sys.exit(1)
 
 

claude_mpm/cli/commands/skills.py

@@ -0,0 +1,434 @@
+"""
+Skills command implementation for claude-mpm.
+
+WHY: This module provides CLI commands for managing Claude Code skills,
+exposing SkillsService functionality for skill discovery, deployment, validation,
+updates, and configuration.
+
+DESIGN DECISIONS:
+- Use BaseCommand pattern for consistency with other CLI commands
+- Rich output formatting for user-friendly display
+- Graceful error handling with informative messages
+- Support for verbose output and structured formats
+"""
+
+import os
+import subprocess
+from typing import Optional
+
+from rich.console import Console
+from rich.markdown import Markdown
+from rich.panel import Panel
+
+from ...constants import SkillsCommands
+from ...skills.skills_service import SkillsService
+from ..shared import BaseCommand, CommandResult
+
+console = Console()
+
+
+class SkillsManagementCommand(BaseCommand):
+    """Skills management command for Claude Code skills."""
+
+    def __init__(self):
+        super().__init__("skills")
+        self._skills_service = None
+
+    @property
+    def skills_service(self) -> SkillsService:
+        """Get skills service instance (lazy loaded)."""
+        if self._skills_service is None:
+            self._skills_service = SkillsService()
+        return self._skills_service
+
+    def validate_args(self, args) -> Optional[str]:
+        """Validate command arguments."""
+        # Most skills commands are optional, basic validation
+        if hasattr(args, "skills_command") and args.skills_command:
+            if args.skills_command == SkillsCommands.VALIDATE.value:
+                if not hasattr(args, "skill_name") or not args.skill_name:
+                    return "Validate command requires a skill name"
+            elif args.skills_command == SkillsCommands.INFO.value:
+                if not hasattr(args, "skill_name") or not args.skill_name:
+                    return "Info command requires a skill name"
+        return None
+
+    def run(self, args) -> CommandResult:
+        """Execute the skills command."""
+        try:
+            # Handle default case (no subcommand) - show list
+            if not hasattr(args, "skills_command") or not args.skills_command:
+                return self._list_skills(args)
+
+            # Route to appropriate subcommand
+            command_map = {
+                SkillsCommands.LIST.value: self._list_skills,
+                SkillsCommands.DEPLOY.value: self._deploy_skills,
+                SkillsCommands.VALIDATE.value: self._validate_skill,
+                SkillsCommands.UPDATE.value: self._update_skills,
+                SkillsCommands.INFO.value: self._show_skill_info,
+                SkillsCommands.CONFIG.value: self._manage_config,
+            }
+
+            handler = command_map.get(args.skills_command)
+            if handler:
+                return handler(args)
+            return CommandResult(
+                success=False,
+                message=f"Unknown skills command: {args.skills_command}",
+                exit_code=1,
+            )
+
+        except Exception as e:
+            self.logger.error(f"Skills command failed: {e}")
+            if hasattr(args, "debug") and args.debug:
+                import traceback
+                traceback.print_exc()
+            return CommandResult(
+                success=False, message=f"Skills command failed: {e}", exit_code=1
+            )
+
+    def _list_skills(self, args) -> CommandResult:
+        """List available skills."""
+        try:
+            # Get skills based on filter
+            if hasattr(args, "agent") and args.agent:
+                skills = self.skills_service.get_skills_for_agent(args.agent)
+                console.print(f"\n[bold cyan]Skills for agent '{args.agent}':[/bold cyan]\n")
+
+                if not skills:
+                    console.print(f"[yellow]No skills found for agent '{args.agent}'[/yellow]")
+                    return CommandResult(success=True, exit_code=0)
+
+                for skill_name in skills:
+                    # Get skill metadata
+                    skill_info = self._get_skill_metadata(skill_name)
+                    if skill_info:
+                        console.print(f"  [green]•[/green] {skill_name}")
+                        if hasattr(args, "verbose") and args.verbose and skill_info.get("description"):
+                            console.print(f"    {skill_info['description']}")
+                    else:
+                        console.print(f"  [green]•[/green] {skill_name}")
+
+            else:
+                # Discover all bundled skills
+                skills = self.skills_service.discover_bundled_skills()
+
+                # Filter by category if specified
+                if hasattr(args, "category") and args.category:
+                    skills = [s for s in skills if s.get("category") == args.category]
+                    console.print(f"\n[bold cyan]Skills in category '{args.category}':[/bold cyan]\n")
+                else:
+                    console.print("\n[bold cyan]Available Skills:[/bold cyan]\n")
+
+                if not skills:
+                    console.print("[yellow]No skills found[/yellow]")
+                    return CommandResult(success=True, exit_code=0)
+
+                # Group by category
+                by_category = {}
+                for skill in skills:
+                    category = skill.get("category", "uncategorized")
+                    if category not in by_category:
+                        by_category[category] = []
+                    by_category[category].append(skill)
+
+                # Display by category
+                for category, category_skills in sorted(by_category.items()):
+                    console.print(f"[bold yellow]{category}[/bold yellow]")
+                    for skill in sorted(category_skills, key=lambda s: s.get("name", "")):
+                        name = skill.get("name", "unknown")
+                        console.print(f"  [green]•[/green] {name}")
+
+                        if hasattr(args, "verbose") and args.verbose:
+                            metadata = skill.get("metadata", {})
+                            if desc := metadata.get("description"):
+                                console.print(f"    {desc}")
+                            if version := metadata.get("version"):
+                                console.print(f"    [dim]Version: {version}[/dim]")
+                    console.print()
+
+            return CommandResult(success=True, exit_code=0)
+
+        except Exception as e:
+            console.print(f"[red]Error listing skills: {e}[/red]")
+            return CommandResult(success=False, message=str(e), exit_code=1)
+
+    def _deploy_skills(self, args) -> CommandResult:
+        """Deploy bundled skills to project."""
+        try:
+            force = getattr(args, "force", False)
+            specific_skills = getattr(args, "skills", None)
+
+            console.print("\n[bold cyan]Deploying skills...[/bold cyan]\n")
+
+            result = self.skills_service.deploy_bundled_skills(
+                force=force,
+                skill_names=specific_skills
+            )
+
+            # Display results
+            if result["deployed"]:
+                console.print(f"[green]✓ Deployed {len(result['deployed'])} skill(s):[/green]")
+                for skill in result["deployed"]:
+                    console.print(f"  • {skill}")
+                console.print()
+
+            if result["skipped"]:
+                console.print(f"[yellow]⊘ Skipped {len(result['skipped'])} skill(s) (already deployed):[/yellow]")
+                for skill in result["skipped"]:
+                    console.print(f"  • {skill}")
+                console.print("[dim]Use --force to redeploy[/dim]\n")
+
+            if result["errors"]:
+                console.print(f"[red]✗ Failed to deploy {len(result['errors'])} skill(s):[/red]")
+                for skill, error in result["errors"].items():
+                    console.print(f"  • {skill}: {error}")
+                console.print()
+
+            # Summary
+            total = len(result["deployed"]) + len(result["skipped"]) + len(result["errors"])
+            console.print(f"[bold]Summary:[/bold] {len(result['deployed'])} deployed, "
+                         f"{len(result['skipped'])} skipped, {len(result['errors'])} errors "
+                         f"(Total: {total})\n")
+
+            # Exit with error if any deployments failed
+            exit_code = 1 if result["errors"] else 0
+            return CommandResult(success=not result["errors"], exit_code=exit_code)
+
+        except Exception as e:
+            console.print(f"[red]Error deploying skills: {e}[/red]")
+            return CommandResult(success=False, message=str(e), exit_code=1)
+
+    def _validate_skill(self, args) -> CommandResult:
+        """Validate skill structure and metadata."""
+        try:
+            skill_name = args.skill_name
+            strict = getattr(args, "strict", False)
+
+            console.print(f"\n[bold cyan]Validating skill '{skill_name}'...[/bold cyan]\n")
+
+            result = self.skills_service.validate_skill(skill_name)
+
+            if result["valid"]:
+                console.print(f"[green]✓ {skill_name} is valid[/green]\n")
+
+                if result.get("warnings"):
+                    console.print(f"[yellow]Warnings ({len(result['warnings'])}):[/yellow]")
+                    for warning in result["warnings"]:
+                        console.print(f"  • {warning}")
+                    console.print()
+
+                    # Treat warnings as errors in strict mode
+                    if strict:
+                        console.print("[red]Strict mode: treating warnings as errors[/red]")
+                        return CommandResult(success=False, exit_code=1)
+
+                return CommandResult(success=True, exit_code=0)
+            console.print(f"[red]✗ {skill_name} has validation errors:[/red]")
+            for error in result.get("errors", []):
+                console.print(f"  • {error}")
+            console.print()
+
+            if result.get("warnings"):
+                console.print("[yellow]Warnings:[/yellow]")
+                for warning in result["warnings"]:
+                    console.print(f"  • {warning}")
+                console.print()
+
+            return CommandResult(success=False, exit_code=1)
+
+        except Exception as e:
+            console.print(f"[red]Error validating skill: {e}[/red]")
+            return CommandResult(success=False, message=str(e), exit_code=1)
+
+    def _update_skills(self, args) -> CommandResult:
+        """Check for and install skill updates."""
+        try:
+            skill_names = getattr(args, "skill_names", [])
+            check_only = getattr(args, "check_only", False)
+            force = getattr(args, "force", False)
+
+            action = "Checking" if check_only else "Updating"
+            console.print(f"\n[bold cyan]{action} skills...[/bold cyan]\n")
+
+            result = self.skills_service.check_for_updates(skill_names)
+
+            if not result.get("updates_available"):
+                console.print("[green]All skills are up to date[/green]\n")
+                return CommandResult(success=True, exit_code=0)
+
+            # Display available updates
+            console.print(f"[yellow]Updates available for {len(result['updates_available'])} skill(s):[/yellow]")
+            for update_info in result["updates_available"]:
+                skill_name = update_info["skill"]
+                current = update_info["current_version"]
+                latest = update_info["latest_version"]
+                console.print(f"  • {skill_name}: {current} → {latest}")
+            console.print()
+
+            if check_only:
+                console.print("[dim]Run without --check-only to install updates[/dim]\n")
+                return CommandResult(success=True, exit_code=0)
+
+            # Install updates
+            console.print("[bold cyan]Installing updates...[/bold cyan]\n")
+            install_result = self.skills_service.install_updates(
+                result["updates_available"], force=force
+            )
+
+            if install_result["updated"]:
+                console.print(f"[green]✓ Updated {len(install_result['updated'])} skill(s)[/green]\n")
+
+            if install_result.get("errors"):
+                console.print(f"[red]✗ Failed to update {len(install_result['errors'])} skill(s)[/red]")
+                for skill, error in install_result["errors"].items():
+                    console.print(f"  • {skill}: {error}")
+                console.print()
+
+            exit_code = 1 if install_result.get("errors") else 0
+            return CommandResult(success=not install_result.get("errors"), exit_code=exit_code)
+
+        except Exception as e:
+            console.print(f"[red]Error updating skills: {e}[/red]")
+            return CommandResult(success=False, message=str(e), exit_code=1)
+
+    def _show_skill_info(self, args) -> CommandResult:
+        """Show detailed skill information."""
+        try:
+            skill_name = args.skill_name
+            show_content = getattr(args, "show_content", False)
+
+            skill_info = self._get_skill_metadata(skill_name)
+
+            if not skill_info:
+                console.print(f"[red]Skill '{skill_name}' not found[/red]")
+                return CommandResult(success=False, exit_code=1)
+
+            # Display skill info in a panel
+            info_text = f"[bold cyan]{skill_name}[/bold cyan]\n\n"
+
+            if desc := skill_info.get("description"):
+                info_text += f"{desc}\n\n"
+
+            if category := skill_info.get("category"):
+                info_text += f"[bold]Category:[/bold] {category}\n"
+
+            if version := skill_info.get("version"):
+                info_text += f"[bold]Version:[/bold] {version}\n"
+
+            if source := skill_info.get("source"):
+                info_text += f"[bold]Source:[/bold] {source}\n"
+
+            # Show agents using this skill
+            agents_using = self.skills_service.get_agents_for_skill(skill_name)
+            if agents_using:
+                info_text += f"\n[bold]Used by agents:[/bold] {', '.join(agents_using)}\n"
+
+            console.print(Panel(info_text, title="Skill Information", border_style="cyan"))
+
+            # Show content if requested
+            if show_content:
+                skill_path = self.skills_service.get_skill_path(skill_name)
+                skill_md = skill_path / "SKILL.md"
+
+                if skill_md.exists():
+                    console.print("\n[bold cyan]Skill Content:[/bold cyan]\n")
+                    content = skill_md.read_text()
+                    console.print(Markdown(content))
+                else:
+                    console.print(f"\n[yellow]SKILL.md not found at {skill_md}[/yellow]")
+
+            return CommandResult(success=True, exit_code=0)
+
+        except Exception as e:
+            console.print(f"[red]Error showing skill info: {e}[/red]")
+            return CommandResult(success=False, message=str(e), exit_code=1)
+
+    def _manage_config(self, args) -> CommandResult:
+        """View or edit skills configuration."""
+        try:
+            scope = getattr(args, "scope", "project")
+            edit = getattr(args, "edit", False)
+            show_path = getattr(args, "path", False)
+
+            config_path = self.skills_service.get_config_path(scope)
+
+            if show_path:
+                console.print(f"\n[cyan]Configuration path ({scope}):[/cyan] {config_path}\n")
+                return CommandResult(success=True, exit_code=0)
+
+            if not config_path.exists():
+                console.print(f"\n[yellow]Configuration file does not exist: {config_path}[/yellow]")
+                console.print("[dim]Would you like to create it? (y/n):[/dim] ", end="")
+
+                if input().lower() == 'y':
+                    self.skills_service.create_default_config(scope)
+                    console.print(f"[green]Created default configuration at {config_path}[/green]\n")
+                else:
+                    return CommandResult(success=False, exit_code=1)
+
+            if edit:
+                # Open in editor
+                editor = os.environ.get("EDITOR", "nano")
+                try:
+                    subprocess.run([editor, str(config_path)], check=True)
+                    console.print(f"\n[green]Configuration saved to {config_path}[/green]\n")
+                    return CommandResult(success=True, exit_code=0)
+                except subprocess.CalledProcessError as e:
+                    console.print(f"[red]Error opening editor: {e}[/red]")
+                    return CommandResult(success=False, exit_code=1)
+            else:
+                # Display config
+                console.print(f"\n[bold cyan]Skills Configuration ({scope}):[/bold cyan]\n")
+                console.print(f"[dim]Path: {config_path}[/dim]\n")
+
+                import yaml
+                config = yaml.safe_load(config_path.read_text())
+                console.print(yaml.dump(config, default_flow_style=False))
+
+                return CommandResult(success=True, exit_code=0)
+
+        except Exception as e:
+            console.print(f"[red]Error managing configuration: {e}[/red]")
+            return CommandResult(success=False, message=str(e), exit_code=1)
+
+    def _get_skill_metadata(self, skill_name: str) -> Optional[dict]:
+        """Get skill metadata from SKILL.md file."""
+        try:
+            skill_path = self.skills_service.get_skill_path(skill_name)
+            skill_md = skill_path / "SKILL.md"
+
+            if not skill_md.exists():
+                return None
+
+            # Parse SKILL.md metadata
+            content = skill_md.read_text()
+            metadata = self.skills_service.parse_skill_metadata(content)
+            return metadata
+
+        except Exception:
+            return None
+
+
+def manage_skills(args) -> int:
+    """
+    Main entry point for skills command.
+
+    Args:
+        args: Parsed command-line arguments
+
+    Returns:
+        Exit code (0 for success, non-zero for failure)
+    """
+    command = SkillsManagementCommand()
+
+    # Validate arguments
+    error = command.validate_args(args)
+    if error:
+        console.print(f"[red]Error: {error}[/red]")
+        return 1
+
+    # Run command
+    result = command.run(args)
+    return result.exit_code

claude_mpm/cli/executor.py

@@ -26,6 +26,7 @@ from .commands import (
 )
 from .commands.analyze_code import manage_analyze_code
 from .commands.dashboard import manage_dashboard
+from .commands.skills import manage_skills
 from .commands.upgrade import upgrade
 
 
@@ -164,6 +165,7 @@ def execute_command(command: str, args) -> int:
         CLICommands.MCP.value: manage_mcp,
         CLICommands.DOCTOR.value: run_doctor,
         CLICommands.UPGRADE.value: upgrade,
+        CLICommands.SKILLS.value: manage_skills,
         "debug": manage_debug,  # Add debug command
         "mpm-init": None,  # Will be handled separately with lazy import
     }

claude_mpm/cli/parsers/base_parser.py

@@ -403,6 +403,13 @@ def create_parser(
     except ImportError:
         pass
 
+    try:
+        from .skills_parser import add_skills_subparser
+
+        add_skills_subparser(subparsers)
+    except ImportError:
+        pass
+
     try:
         from .config_parser import add_config_subparser