Project Resources in Kaddo: connecting real system context
Project Resources make the external dependencies that shape a system visible, so people and agents can connect modules, capabilities and Work Items to real operating context.
Kaddo Blog
Technical articles about Knowledge-Driven Development, AI-assisted software engineering, software architecture, agents and evolving software systems.
Featured
What is Knowledge-Driven Development?
Knowledge-Driven Development treats project knowledge as a first-class, observable artifact so humans and AI agents evolve software from the same shared context.
Diagrams as knowledge: Mermaid in Business, Product and Tech
Kaddo now treats Mermaid diagrams as first-class knowledge in Business, Product and Tech. The source stays in Markdown; the Admin renders it — architecture, flows and actors you can actually see.
Giving AI context is not enough: AI-assisted development needs a full lifecycle
Giving an AI agent context does not guarantee a correct result. AI-assisted development needs a full lifecycle that connects intent, planning, implementation, evidence, verification and learning.
Latest
Project Resources in Kaddo: connecting real system context
Project Resources make the external dependencies that shape a system visible, so people and agents can connect modules, capabilities and Work Items to real operating context.
How to work with legacy projects using AI without breaking what already works
AI does not remove the uncertainty of a legacy system by itself. A safe approach makes risks, unknowns and dependencies visible before changing code.
What is Knowledge-Driven Development?
Knowledge-Driven Development treats project knowledge as a first-class, observable artifact so humans and AI agents evolve software from the same shared context.
Diagrams as knowledge: Mermaid in Business, Product and Tech
Kaddo now treats Mermaid diagrams as first-class knowledge in Business, Product and Tech. The source stays in Markdown; the Admin renders it — architecture, flows and actors you can actually see.
Giving AI context is not enough: AI-assisted development needs a full lifecycle
Giving an AI agent context does not guarantee a correct result. AI-assisted development needs a full lifecycle that connects intent, planning, implementation, evidence, verification and learning.
How Kiro Powers ease the adoption of Knowledge-Driven Development with Kaddo
Having the capabilities is not enough; adoption has to be easy too. A Kiro Power activates Kaddo's MCP server, Skills and KDD flow right inside the agentic IDE.
For agents, not every task needs the same knowledge
The knowledge an AI agent needs should grow with the scope of the task. The challenge is not giving it all the information possible, but the right information.
Not all projects develop with AI the same way
New, pre-AI and legacy projects need different AI strategies. Before talking about AI-assisted development, we have to talk about the real state of the project and the knowledge the AI needs.