feat: Add Digital Product Manager Agent with PRD and Roadmap capabilities

[criptogus]

📋 Description

This PR introduces a new Digital Product Manager Agent - a specialized AI agent with comprehensive expertise in digital product management. The agent helps users create Product Requirements Documents (PRDs), product roadmaps, and provides strategic guidance across product management, UX/UI design, technology architecture, AI/ML integration, marketing, and business strategy.

Key Features:

• PRD Generation: Creates professional, complete Product Requirements Documents
• Roadmap Planning: Generates strategic product roadmaps with prioritization frameworks (RICE, MoSCoW, ICE, Kano)
• Product-Market Fit: Guidance on ICP, Value Proposition Canvas, JTBD, TAM/SAM/SOM
• UX/UI Expertise: Nielsen heuristics, Design Systems, Information Architecture, Accessibility (WCAG)
• Technology & Architecture: System design, API design, database design, cloud architecture, DevOps, security
• AI/ML Integration: AI-powered features, LLM integration, RAG, prompt engineering, AI UX patterns
• Marketing & Growth: Growth loops, SEO/SEM, funnel analysis, analytics, A/B testing, PLG
• Business Strategy: Unit economics, pricing strategy, go-to-market, OKRs, competitive analysis

Implementation Details:

• New agent configuration in intelligent-router.ts with proper tool bindings
• Comprehensive system prompt in specialized-agent-factory.ts with detailed competencies and rules
• Dedicated prompt file digital-product-agent.ts with enterprise-grade documentation
• Integration with existing tools: generate_prd, generate_product_roadmap, web_search, memory management, task/goal creation
• Uses Gemini Pro model for complex analysis and document generation

🎯 Type of Change

• ✨ New feature (non-breaking change which adds functionality)

🔗 Related Issues

Related to product management and digital product strategy capabilities expansion.

✅ Checklist

Code Quality

• My code follows the project's style guidelines
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• My changes generate no new warnings or errors
• I have removed console.logs and debug code

Testing

• Code follows established patterns from other agents (advisory_board, james_clear_habits, podcast)
• Agent configuration properly integrated into router
• Tool bindings verified against existing tool implementations

Documentation

• Comprehensive system prompt with detailed competencies
• Clear rules and guidelines for agent behavior
• Example queries provided for user guidance
• Framework references documented

Dependencies

• No new dependencies added
• Uses existing tools and infrastructure

Security

• No sensitive information exposed
• Follows security guidelines from other agents
• Proper tool access control through agent configuration

🧪 How Has This Been Tested?

• Code structure validated against existing agent patterns
• Tool bindings verified against available tools in the system
• Prompt structure follows established guidelines from other specialized agents
• Configuration schema compliance verified

📝 Additional Notes

The Digital Product Manager Agent is designed to be:

• Practical & Actionable: Provides specific, contextual advice rather than generic guidance
• Holistic: Considers technical, UX, and business implications of recommendations
• Enterprise-Grade: Generates professional-quality documents and deliverables
• Resilient: Continues providing value even if individual tools fail
• Proactive: Identifies risks, dependencies, and suggests KPIs for success

The agent leverages multiple frameworks (Design Thinking, Lean Startup, Shape Up, RICE, MoSCoW, etc.) and can adapt recommendations based on user context and project history.

🚀 Deployment Notes

• No database migrations required
• No environment variables need to be updated
• No configuration changes needed beyond agent registration
• No new dependencies to install

✨ Future Improvements

https://claude.ai/code/session_01Kd8ygVGieU1a84u5QZmdWd

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.