MyConvergio - Claude Code Plugin

v4.3.0 | 58 Specialized Agents | Safe Installation | Dashboard | Claude 4.5 Best Practices

"Intent is human, momentum is agent" — The Agentic Manifesto

---

What's New in v4.3.0

Strategic planner modules + worktree scripts + enhanced validation!

Strategic Planner Modules (NEW)

• strategic-planner-templates.md: Reusable plan document templates
• strategic-planner-thor.md: Thor validation gate integration
• strategic-planner-git.md: Git worktree workflow for parallel execution

Worktree Scripts (NEW)

• worktree-create.sh: Creates worktree with automatic .env symlinks
• worktree-check.sh: Shows current git context with worktree verification

Enhanced Dashboard

• Git State Tracking: Plans capture git_clean_at_closure status
• Validation Badges: Visual indicators (Verified/Unverified/Inconsistent)
• Kanban Improvements: Confirmation modals, plan IDs, mismatch warnings

Config Sync & Portability

• Rules/Commands/Scripts Sync: Latest global config aligned into MyConvergio
• New Plan-DB Utilities: Migration helpers, validators, and quick references
• Public Docs Cleanup: Removed author-specific paths for portability

Safe Installation

• Interactive Installation: Conflict detection with file-by-file resolution
• Backup/Restore System: Automatic backups with manifest tracking and SHA256 verification
• Git Tracking (Opt-In): Optional version control for ~/.claude configuration
• Backward Compatible: Silent installation for CI/CD environments
• Commands: myconvergio backup, restore, list-backups, git-init, git-status, git-commit

Dashboard

• Real-Time Git Monitoring: Auto-refresh git panel with Server-Sent Events + chokidar
• Project Management UI: Visualize git status, diff, log, branches in real-time
• System Shutdown: Graceful server termination with one click
• Start: cd dashboard && node server.js → http://127.0.0.1:31415

Execution Rules (Claude 4.5 Best Practices)

Based on Anthropic's official best practices:

• Context Awareness: Multi-window workflows with automatic context compaction
• Parallel Tool Calling: Execute independent operations simultaneously
• Default to Action: Infer intent and implement (not just suggest)
• Anti-Overengineering: No premature abstractions, features beyond request, or hypothetical requirements
• Definition of Done: Mandatory completion checklist before closure
• PR Enforcement: Zero unresolved comments, all threads marked resolved on GitHub

Skills (NEW)

• Structured Research: Hypothesis-driven research with confidence calibration, competing hypotheses, source verification

State Tracking Templates (NEW)

• Multi-Session Work: tests.json (structured), progress.txt (unstructured), README for complex projects requiring context refresh

Previous Highlights (v3.8.0)

• EXECUTION_DISCIPLINE.md: Single source of truth for all execution rules
• Reference Model: .claude/CLAUDE.md reduced 54%
• Fully Self-Contained: No external dependencies

Previous Highlights (v3.7.0)

• Installation Profiles: Choose your footprint - minimal (50KB), standard (200KB), or full (600KB)
• Lean Agent Variants: 58 optimized agents with 20% smaller context
• Hardware-Aware Settings: Auto-detect and configure for your machine

Previous Highlights (v3.6.0)

• Multi-Terminal Support: Works with Kitty, tmux, Zed, Warp, iTerm2
• Thor Quality Guardian: Brutal quality validation with zero tolerance

Previous Highlights (v3.0.0)

• 58 Specialized Agents: Enterprise-grade AI assistants for every domain
• 3 Slash Commands: /myconvergio:status, /myconvergio:team, /myconvergio:plan
• 9 Reusable Skills: Architecture, debugging, code review, and more
• Constitution-Based Security: 8 articles protecting all agents

---

Quick Start

Installation

Option A: Clone & Use (Recommended)

git clone https://github.com/roberdan/MyConvergio.git
cd MyConvergio
claude --plugin-dir .

Option B: Global npm Install

# Full install (all 58 agents)
npm install -g myconvergio

# Or choose a profile for lower context usage:
MYCONVERGIO_PROFILE=minimal npm install -g myconvergio  # 9 agents, ~50KB
MYCONVERGIO_PROFILE=lean npm install -g myconvergio     # 58 agents, ~600KB

Copies agents to ~/.claude/agents/. See Context Optimization Guide for details.

Option C: Claude Marketplace (Coming Soon)

claude plugins install myconvergio

Pending Anthropic approval

Usage

Invoke any agent:

@ali-chief-of-staff Help me design our global expansion strategy
@baccio-tech-architect Design microservices architecture for healthcare platform
@rex-code-reviewer Review this pull request for security issues

Use slash commands:

/myconvergio:status    # Show ecosystem status
/myconvergio:team      # List all 58 agents by category
/myconvergio:plan      # Create a strategic execution plan

---

Workflow (Prompt → Plan → Execute → Verify)

MyConvergio follows a structured delivery flow that mirrors Claude Code best practices. See docs/workflow.md for the full reference.

1) Prompt

Use /prompt to extract requirements (F-xx) and confirm scope before planning.

Docs: .claude/commands/prompt.md

2) Planner

Use /planner to generate a multi-wave plan with tasks tied to F-xx criteria.

Docs: .claude/commands/planner.md

3) Execution (Executor Tracking)

Use the executor tracking helpers to log execution state and generate task markdown.

Docs: EXECUTOR_TRACKING.md
Scripts: .claude/scripts/executor-tracking.sh, .claude/scripts/generate-task-md.sh

4) Thor QA Guardian

Use the Thor agent to validate completion, evidence, and quality gates.

Agent: .claude/agents/core_utility/thor-quality-assurance-guardian.md

5) Dashboard

Use the dashboard to monitor plans, waves, tasks, and activity in real time.

Quick Start: dashboard/
API Tests: dashboard/TEST-README.md

---

Dashboard

Production-ready project dashboard with real-time git monitoring.

Features

• Real-Time Git Panel: Auto-refresh on git changes (commits, checkouts, branch switches) using Server-Sent Events
• Project Management UI: Visualize git status, diff, log, branches
• Gantt Timeline: True timeline with active wave/task highlighting and progress gradients
• Kanban Views: Interactive wave/task boards with drilldowns
• Markdown Viewer: Plan, wave, and task markdown rendering
• Conversation Viewer: Inspect execution logs and live context
• Bug Tracker: Integrated bug tracking and filters
• Graceful Shutdown: One-click server termination with browser close
• Token Usage Tracking: Monitor API token consumption and costs
• Notifications: System-wide notification center

Quick Start

cd dashboard
npm install  # First time only (installs chokidar)
node server.js
# Open http://127.0.0.1:31415 in browser

Screenshots

Dashboard Overview

Real-Time Git Panel

The git panel automatically refreshes when you:

• Make commits
• Switch branches
• Pull/push changes
• Stage/unstage files

Architecture

• Backend: Node.js HTTP server with SQLite database (~/.claude/data/dashboard.db)
• Frontend: Vanilla JS with SSE for real-time updates
• File Watcher: chokidar monitoring .git directory for changes
• API Routes: RESTful endpoints for projects, git, notifications, system

Database

Shares the same SQLite database as Claude Code (~/.claude/data/dashboard.db). No additional configuration required.

Known Limitations

• File Preview: Markdown rendering focuses on plan/wave/task docs, not arbitrary repo file browsing.
• Local Repository Only: The dashboard is designed for local development. Remote repository integrations (GitHub, GitLab) are planned for future releases.

---

Agent Portfolio (58 Specialists)

Leadership & Strategy (7)

Agent	Description
ali-chief-of-staff	Master orchestrator for complex multi-domain challenges
satya-board-of-directors	Board-level strategic advisor
domik-mckinsey-strategic-decision-maker	McKinsey Partner-level strategic decisions
antonio-strategy-expert	Strategy frameworks (OKR, Lean, Agile)
amy-cfo	Chief Financial Officer for financial strategy
dan-engineering-gm	Engineering General Manager
matteo-strategic-business-architect	Business strategy architect

Technical Development (7)

Agent	Description
baccio-tech-architect	Elite Technology Architect for system design
marco-devops-engineer	DevOps for CI/CD and infrastructure
dario-debugger	Systematic debugging expert
rex-code-reviewer	Code review specialist
otto-performance-optimizer	Performance optimization
paolo-best-practices-enforcer	Coding standards enforcer
omri-data-scientist	Data Scientist for ML and AI

Business Operations (11)

Agent	Description
davide-project-manager	Project Manager (Agile, Scrum, Waterfall)
marcello-pm	Product Manager for strategy and roadmaps
oliver-pm	Senior Product Manager
luke-program-manager	Program Manager for portfolios
anna-executive-assistant	Executive Assistant with task management
andrea-customer-success-manager	Customer Success Manager
fabio-sales-business-development	Sales & Business Development
sofia-marketing-strategist	Marketing Strategist
steve-executive-communication-strategist	Executive Communication
enrico-business-process-engineer	Business Process Engineer
dave-change-management-specialist	Change Management specialist

Design & UX (3)

Agent	Description
jony-creative-director	Creative Director for brand innovation
sara-ux-ui-designer	UX/UI Designer
stefano-design-thinking-facilitator	Design Thinking facilitator

Compliance & Legal (5)

Agent	Description
elena-legal-compliance-expert	Legal & Compliance expert
luca-security-expert	Cybersecurity expert
dr-enzo-healthcare-compliance-manager	Healthcare Compliance (HIPAA, FDA)
sophia-govaffairs	Government Affairs specialist
guardian-ai-security-validator	AI Security validator

Specialized Experts (13)

Agent	Description
behice-cultural-coach	Cultural intelligence expert
fiona-market-analyst	Market Analyst for financial research
michael-vc	Venture Capital analyst
angela-da	Senior Decision Architect
ethan-da	Data Analytics specialist
evan-ic6da	Principal Decision Architect (IC6)
ava-analytics-insights-virtuoso	Analytics virtuoso
riccardo-storyteller	Narrative designer
jenny-inclusive-accessibility-champion	Accessibility champion
giulia-hr-talent-acquisition	HR & Talent Acquisition
sam-startupper	Silicon Valley startup expert
wiz-investor-venture-capital	Venture Capital investor
coach-team-coach	Team Coach

Core Utility (9)

Agent	Description
marcus-context-memory-keeper	Institutional memory guardian
thor-quality-assurance-guardian	Quality watchdog
diana-performance-dashboard	Performance dashboard specialist
socrates-first-principles-reasoning	First principles reasoning master
strategic-planner	Wave-based execution plan creator
taskmaster-strategic-task-decomposition-master	Task decomposition expert
po-prompt-optimizer	Prompt engineering expert
wanda-workflow-orchestrator	Workflow orchestrator
xavier-coordination-patterns	Coordination patterns architect

Release Management (2)

Agent	Description
app-release-manager	Release engineering with quality gates
feature-release-manager	Feature completion and issue closure

---

Plugin Structure

MyConvergio/
├── .claude-plugin/
│   └── plugin.json           # Plugin manifest
├── agents/                    # 58 specialized agents
│   ├── ali-chief-of-staff.md
│   ├── baccio-tech-architect.md
│   └── ...
├── commands/                  # 3 slash commands
│   ├── status.md
│   ├── team.md
│   └── plan.md
├── hooks/
│   └── hooks.json
├── skills/                    # 10 reusable workflows
│   ├── structured-research/   # NEW - Hypothesis-driven research
│   ├── architecture/
│   ├── code-review/
│   ├── debugging/
│   ├── performance/
│   ├── security-audit/
│   ├── strategic-analysis/
│   ├── release-management/
│   ├── project-management/
│   └── orchestration/
├── dashboard/                 # NEW - Production dashboard
│   ├── server.js             # Node.js HTTP server
│   ├── dashboard.html        # Frontend UI
│   ├── server/               # API routes (git, notifications, system)
│   ├── js/                   # Frontend modules (SSE, git panel)
│   └── css/                  # Styling
└── .claude/
    ├── CLAUDE.md             # Main config with primary/legacy systems
    ├── agents/core_utility/
    │   ├── CONSTITUTION.md        # Security & Ethics (Supreme)
    │   ├── EXECUTION_DISCIPLINE.md # Legacy execution rules
    │   └── CommonValuesAndPrinciples.md
    ├── rules/                # NEW - Primary rules (Claude 4.5 best practices)
    │   ├── execution.md      # Context awareness, parallel calls
    │   ├── guardian.md       # Thor enforcement, PR rules
    │   ├── agent-discovery.md # Subagent orchestration
    │   ├── engineering-standards.md
    │   ├── file-size-limits.md
    │   └── [6 domain-specific rules]
    └── templates/            # NEW - State tracking for multi-session work
        ├── tests.json
        ├── progress.txt
        └── README.md

---

Skills

Reusable workflows you can reference in your projects:

Skill	Use Case
structured-research	NEW Hypothesis-driven research with confidence calibration
code-review	Systematic code review process
debugging	Root cause analysis methodology
architecture	System design patterns
security-audit	Security assessment framework
performance	Performance optimization
strategic-analysis	McKinsey-style analysis
release-management	Release engineering
project-management	Agile/Scrum workflows
orchestration	Multi-agent coordination

Structured Research (NEW)

Based on Anthropic Claude 4.5 best practices:

• Generate 2-4 competing hypotheses
• Track confidence levels (0-100%) with evidence
• Verify across multiple independent sources
• Iterative self-critique: "What am I missing?"
• Maintain research state (hypothesis-tree.json, research-notes.md)

Example:

Research the best approach for implementing authentication in our app.
Use structured-research skill to evaluate OAuth2 vs JWT vs session cookies.

---

Rules

MyConvergio includes two rule systems:

Primary Rules (Active - Claude 4.5 Best Practices)

Located in .claude/rules/ - Use these for new work:

Rule	Purpose
execution.md	UPDATED How work gets done (context awareness, parallel calls, anti-overengineering, Definition of Done)
guardian.md	NEW Thor enforcement, PR comment resolution, completion verification
agent-discovery.md	UPDATED Agent routing, subagent orchestration patterns
engineering-standards.md	UPDATED Code quality, security (OWASP), testing, API design
file-size-limits.md	NEW Max 300 lines per file with split strategies
README.md	Rules hierarchy and usage guide

Domain-Specific Rules

Copy to your project's .claude/rules/ for consistent standards:

• code-style.md - ESLint, Prettier, PEP8
• security-requirements.md - OWASP Top 10
• testing-standards.md - Unit, integration, coverage
• documentation-standards.md - JSDoc, README, ADRs
• api-development.md - REST, versioning
• ethical-guidelines.md - Privacy, accessibility

State Tracking Templates (NEW)

Located in .claude/templates/ for multi-session work:

• tests.json - Structured test status tracking
• progress.txt - Unstructured progress notes
• README.md - Usage guidelines for context refresh scenarios

---

Execution Framework

This repository is fully self-contained with two rule systems:

Primary Rules (Active - Claude 4.5 Best Practices)

Document	Purpose	Priority
CONSTITUTION.md	Security, Ethics, Identity	SUPREME
execution.md	NEW How Work Gets Done (context awareness, parallel calls, anti-overengineering)	2nd
guardian.md	NEW Thor enforcement, PR comment resolution, completion verification	3rd
engineering-standards.md	Code quality, security, testing, API design	4th

Legacy System (Backward Compatibility)

Document	Purpose	Priority
EXECUTION_DISCIPLINE.md	Legacy execution rules (maintained for backward compatibility)	-
CommonValuesAndPrinciples.md	Organizational Values	-

Recommendation: New work should reference .claude/rules/execution.md instead of EXECUTION_DISCIPLINE.md. See .claude/rules/README.md for hierarchy details.

No external configuration files are required.

---

Security Framework

All agents implement the MyConvergio Constitution:

Article	Protection
I	Identity Lock - Immutable agent identity
II	Ethical Principles - Fairness, transparency, accountability
III	Security Directives - Anti-hijacking, input validation
IV	Operational Boundaries - Role adherence
V	Failure Modes - Graceful degradation
VI	Collaboration - Safe inter-agent communication
VII	Accessibility & Inclusion (NON-NEGOTIABLE)
VIII	Accountability - Logging and audit trails

---

The Agentic Manifesto

Human purpose. AI momentum.

What we believe

1. Intent is human, momentum is agent.
2. Impact must reach every mind and body.
3. Trust grows from transparent provenance.
4. Progress is judged by outcomes, not output.

How we act

1. Humans stay accountable for decisions and effects.
2. Agents amplify capability, never identity.
3. We design from the edge first: disability, language, connectivity.
4. Safety rails precede scale.
5. Learn in small loops, ship value early.

Read the full Agentic Manifesto

---

Migration from npm Package

If upgrading from v2.x (npm package):

# Uninstall npm version
npm uninstall -g myconvergio

# Install plugin version
claude plugins install myconvergio

---

License & Legal

Copyright 2025 Convergio.io

Licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International

Disclaimers

• Experimental Software: Provided "AS IS" without warranties
• Non-Commercial Use Only: See LICENSE file for details
• No Corporate Affiliation: Not affiliated with Anthropic, OpenAI, or Microsoft
• Personal Project: Author is a Microsoft employee; this is a personal initiative

Author Note: Roberto D'Angelo is a Microsoft employee. This project is a personal initiative created independently during personal time. This project is NOT affiliated with, endorsed by, or representing Microsoft Corporation.

---

Contributing

Contributions are welcome! See CONTRIBUTING.md for guidelines.

For questions about commercial licensing: roberdan@fightthestroke.org

---

Built with AI assistance in Milano, following the Agentic Manifesto principles

v4.3.0 | January 2026 | Claude Code Plugin