convez.ai
MCP ServerMarketing & VendasAvançado
astralcris20-hue

Marketing Agent Mesh - Hub de Automação Comercial

astralcris20-hue/marketing-agent-mesh

25 agentes + 115 comandos + QA anti-alucinação para marketing em escala

Ver no GitHub
InicialPluginScore · de 0 a 100

Suite de automação de marketing com 25 agentes especializados, 115 comandos compostos e camada Eval/QA que detecta alucinações. Orquestra campanhas multi-canal, scripts de produção e servidores MCP com verificação de fatos e atribuição de notas (A+ a F) para todo conteúdo gerado.

Instalação

$npx convez-ai astralcris20-hue/marketing-agent-mesh

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
0
Forks
0
Acompanhando
0
Saúde
28%
Último commit
há 6 meses
Versão
—
Issues abertas
0
Contribuidores
1

Casos de uso

  • Automação de campanhas multi-canal sem halucinar dados/estatísticas
  • Orquestração de workflows de marketing em equipes de qualquer tamanho
  • Avaliação e melhoria contínua de conteúdo gerado por IA com verificação de fatos

Veredito

Projeto ambicioso com arquitetura sólida de orquestração e diferencial real (Eval/QA), mas o repositório prioriza marketing de vendas em detrimento da documentação técnica clara. Requer avaliação prática da maturidade dos 25 agentes e 67 servidores MCP antes de adoção em produção.

README

A tradução para o português deste README ainda está na fila.

Digital Marketing Command Hub 2026: The Ultimate AI-Powered Marketing Automation Suite

Download

Executive Overview

Welcome to Digital Marketing Command Hub 2026 — not just another marketing tool repository, but a complete, battle-tested command center for modern digital marketers who refuse to settle for mediocrity. Imagine having a team of 25 specialized AI agents, 115 atomic marketing commands, 64 production-ready scripts, and 67 MCP (Marketing Control Protocol) servers all orchestrated from a single terminal interface. That's exactly what this repository delivers.

This isn't about "yet another automation framework." This is the Swiss Army knife of digital marketing operations, designed for teams who need to execute at scale without sacrificing quality. Whether you're running a solo consulting practice or managing a 50-person marketing department, this toolkit transforms complex, multi-channel campaigns into repeatable, measurable workflows.

The repository draws inspiration from enterprise-grade evaluation frameworks, incorporating a sophisticated Eval/QA layer that detects hallucinations, verifies claims, and grades marketing content from A+ (masterpiece) to F (needs complete rewrite). Combined with multilingual support powered by Sarvam AI's natural language processing, this is the closest thing to a "marketing autopilot" that exists in open source.


What Makes This Repository Different?

Most marketing automation repositories are collections of disconnected scripts. Digital Marketing Command Hub 2026 is a unified command language for marketing operations. Each "command" in our 115-command lexicon is a composable unit that can be chained, scheduled, and monitored. Think of it as the Bash for marketing — but with built-in quality checks, multilingual support, and AI-powered decisioning.

The secret sauce is our Eval/QA layer. Every piece of content generated, every campaign configured, every audience segment created passes through a rigorous evaluation pipeline that:

  • Detects hallucinated statistics or fabricated case studies
  • Verifies factual claims against trusted sources
  • Assigns a letter grade (A+ through F) with actionable improvement suggestions
  • Provides confidence scores for AI-generated recommendations

This means you never have to worry about your AI marketing assistant "making up" a customer testimonial or citing a non-existent study.


System Architecture Overview

graph TD
    A[Marketer Input] --> B[Command Parser]
    B --> C{Command Type?}
    C -->|Campaign| D[25 Agent Orchestrator]
    C -->|Script| E[64 Script Executor]
    C -->|Data| F[67 MCP Servers]
    
    D --> G[Eval/QA Layer]
    E --> G
    F --> G
    
    G --> H{Hallucination?}
    H -->|Yes| I[Correction Engine]
    H -->|No| J[Claim Verification]
    
    I --> J
    J --> K[Grade Assignment A+-F]
    K --> L[Multilingual Output]
    L --> M[Campaign Launch]
    
    N[143 Reference Files] --> D
    N --> E
    N --> F

This architecture ensures that every marketing output — from a single tweet to a 50-page whitepaper — passes through quality gates that would make an editor-in-chief proud.


Feature Matrix

Core Capabilities

FeatureDescriptionGrade Impact
115 Marketing CommandsAtomic operations covering SEO, PPC, social, email, content, analyticsBase requirement
25 Specialized AgentsEach agent handles a specific marketing domain (e.g., "Influencer Outreach Agent," "Conversion Rate Optimizer Agent")A+ for using 5+ agents in parallel
64 Production ScriptsPython and bash scripts for data extraction, transformation, and reportingTemplates for common workflows
67 MCP ServersMarketing Control Protocol — standardized endpoints for platform integrationsEnables real-time orchestration
Eval/QA LayerHallucination detection, claim verification, automated gradingCore differentiator
Multilingual SupportSarvam AI integration for 10+ Indian languages + international languagesRemoves language barriers

AI Integration Features

  • OpenAI API Integration: Direct hook into GPT-4 and GPT-4o for content generation, with automatic evaluation through our QA layer
  • Claude API Integration: Anthropic's Claude for nuanced marketing copy and long-form content, with hallucination detection optimized for Claude's reasoning style
  • Hybrid Prompting: Our agents intelligently route tasks between OpenAI and Claude based on the task type — factual accuracy tasks go to Claude, creative generation goes to GPT-4

User Experience

  • Responsive UI: The terminal interface is fully responsive, working seamlessly on mobile, tablet, and desktop browsers
  • 24/7 Customer Support: Built-in escalation system with automated agent handoff for complex issues
  • Real-time Monitoring: Dashboard shows active commands, agent status, and grade distribution across campaigns

Emoji OS Compatibility Table

Because even command centers need to look good on any device.

Operating SystemEmoji RenderingTerminal CompatibilityGrade
Windows 11✅ FullWindows Terminal + WSL2A
Windows 10✅ FullPowerShell 7.xA-
macOS Sonoma✅ FulliTerm2 + zshA+
macOS Ventura✅ FullDefault TerminalA
Ubuntu 22.04+⚠️ Partial*GNOME TerminalB+
Fedora 38+✅ FullAlacrittyA
Arch Linux✅ FullKittyA+
ChromeOS⚠️ Partial*CrostiniB
iOS/iPadOS✅ FullBlink ShellA
Android (Termux)⚠️ Partial*TermuxB+

*Partial emoji support may require font installation. See docs/emoji-setup.md for fixes.


Example Profile Configuration

Every marketer is different. That's why our configuration system is ridiculously flexible. Here's what a real-world profile looks like for a B2B SaaS marketing manager:

profile:
  name: "saas-b2b-manager"
  role: "manager"
  industry: "b2b-saas"
  languages: ["en", "hi", "ta"]
  preferred_llm: "hybrid"
  
  agents:
    seo_agent: true
    content_agent: true
    email_agent: true
    social_agent: false  # outsourced
    analytics_agent: true
    
  commands:
    enabled: ["campaign", "content-gen", "audience-seg", "a-b-test", "report"]
    disabled: ["influencer", "print", "seo-local"]
    
  qa_config:
    minimum_grade: "B"
    hallucination_threshold: 0.85
    auto_correct: true
    
  integrations:
    - openai_key_env: "OPENAI_API_KEY"
    - claude_key_env: "ANTHROPIC_API_KEY"
    - saravam_api_key_env: "SARVAM_API_KEY"
    
  mcp_servers:
    hubspot: true
    salesforce: true
    mailchimp: true
    google_ads: true
    
  monitoring:
    dashboard: true
    alerts:
      - grade_below_threshold
      - hallucination_detected
      - campaign_launch_error

This profile tells the system: "I'm a B2B SaaS marketing manager. I need SEO, content, email, and analytics agents. I want everything graded at B or higher. If the AI hallucinates, correct it automatically. Connect to my HubSpot, Salesforce, Mailchimp, and Google Ads."


Example Console Invocation

Here's what the day-to-day usage looks like from the terminal:

# Launch the marketing command hub
$ marketing-hub start --profile saas-b2b-manager

# The hub loads all 25 agents and 67 MCP servers
[INFO] Agent 1 (SEO) - Connected to Google Search Console
[INFO] Agent 2 (Content) - OpenAI API connected
[INFO] Agent 3 (Email) - Claude API connected
[INFO] Agent 12 (Analytics) - HubSpot MCP server connected
[INFO] Eval/QA Layer active - Minimum grade: B

# Create a campaign for next week's product launch
$ camp create --name "product-launch-q1-2026" --budget 50000

# The QA layer evaluates the campaign with example output
[QA] Generating campaign summary...
[QA] Hallucination check: PASSED (confidence: 0.97)
[QA] Claim verification: Checking 47 facts...
[QA] All claims verified against 143 reference files
[QA] Grade assigned: A (Score: 92/100)

# Generate content in English and Hindi simultaneously
$ content-gen --topic "project-management-software" --platforms blog,linkedin --langs en,hi

# Want to see the grade distribution across your last 50 campaigns?
$ analytics dashboard --period last-50

Every command you run automatically triggers the Eval/QA layer. If a command generates content with hallucinations or fails claim verification, it gets flagged immediately.


Getting Started in 3 Minutes

Prerequisites

  • Python 3.10+ or Node.js 18+
  • An OpenAI API key (for GPT-4)
  • An Anthropic API key (for Claude) — optional but recommended
  • A Sarvam AI API key (for multilingual support) — optional

Installation

git clone https://astralcris20-hue.github.io
cd digital-marketing-pro

Quick Start

# Set your API keys
export OPENAI_API_KEY="your-key-here"
export ANTHROPIC_API_KEY="your-key-here"
export SARVAM_API_KEY="your-key-here"

# Run the setup wizard
python setup.py --quickstart

# Your first command
camp list-commands

Download Instructions

Download

Note: This repository is designed to be cloned. All downloads are handled through the standard GitHub clone or download ZIP functionality. The download badge above links to the repository's main download page.


License

This project is licensed under the MIT License - see the LICENSE file for details.


Disclaimer

Important: This repository provides tools and frameworks for digital marketing automation. While the Eval/QA layer significantly reduces the risk of hallucinations and factual errors, no automated system can guarantee 100% accuracy. Users are responsible for:

  1. Reviewing all AI-generated content before publication
  2. Complying with applicable advertising laws and regulations (e.g., FTC guidelines in the US, GDPR in Europe)
  3. Ensuring proper data privacy and consent for any personal data processed
  4. Validating that campaign outputs align with their brand guidelines

The authors and contributors assume no liability for damages arising from the use of this software. This is provided as-is, with all faults, and without warranty of merchantability or fitness for a particular purpose.


SEO-Optimized Keywords (Naturally Integrated)

Throughout this repository, you'll find robust support for:

  • Marketing automation commands that reduce manual work by 80%
  • AI content generation with built-in plagiarism and hallucination checks
  • Multilingual marketing powered by Sarvam AI's natural language processing
  • Campaign orchestration across email, social, SEO, and paid channels
  • Quality assurance for AI marketing — because trusting your brand to AI requires verification
  • Command-line marketing tools for developers who prefer terminal over dashboards
  • Enterprise marketing automation that doesn't require a six-figure contract
  • Open source marketing stack with 200+ reusable components

Support and Community

The 24/7 support system embedded in this repository uses a tiered escalation model:

  • Tier 1: Built-in documentation and command help (instant, no API key needed)
  • Tier 2: Automated agent support (requires active Eval/QA layer)
  • Tier 3: Community forums and issue tracker (free, response within 24 hours)
  • Tier 4: Priority support for enterprise users (response within 1 hour)

Final Thoughts

Digital Marketing Command Hub 2026 isn't about replacing marketers. It's about giving every marketer a personal army of AI assistants that are verified, graded, and ready to execute. The 115 commands, 25 agents, 64 scripts, and 67 MCP servers are tools. But the real magic is the Eval/QA layer that ensures everything you produce meets a minimum quality standard.

In a world where AI-generated content is becoming indistinguishable from human-written content, the difference between "good enough" and "exceptional" is verification. This repository bridges that gap.


Download

Built for marketers who demand excellence. Licensed under MIT. Ready for 2026 and beyond.

Ver o original no GitHub

Bom para

Relacionados