AI agent engineering guides for people building real agents

Practical pages on Eve, MCP, tools, skills, subagents, durable runs, shadcn registries, and framework tradeoffs. Start with the decision, then inspect installable source when you are ready to build.

Protocols

MCP, skills, tools, resources, and how agents connect to external systems.

MCP server for AI agentsA practical guide to MCP servers for agent builders: tools, resources, prompts, permissions, and where MCP stops.MCP vs skillsA grounded distinction between MCP servers, tools, resources, prompts, and skills for teams extending AI agents.

Agent engineering

Production choices behind workflows, tools, approval, state, and recovery.

Agentic workflowsHow to decide when a workflow should be agentic, deterministic, or a hybrid of both.AI agent toolsHow to design agent tools with narrow authority, typed inputs, clear descriptions, and observable results.

Comparisons

Framework tradeoffs for teams choosing how to build and operate agents.

AI agent frameworksA practical checklist for comparing LangGraph, CrewAI, AutoGen-style systems, Eve, and other agent frameworks.LangGraph vs CrewAIA builder-focused comparison of LangGraph and CrewAI, with Eve as a file-based reference point.

Distribution

Registries, source-owned agents, and installable workflow files.

Agent registryWhat an AI agent registry must expose before developers can reuse agents safely.Shadcn registry for agentsHow the shadcn registry model applies to AI agent workflows, where users need source files rather than opaque packages.

All guides

8 indexable guides, grouped by practical agent decisions.

MCP server for AI agents

A practical guide to MCP servers for agent builders: tools, resources, prompts, permissions, and where MCP stops.

Read guide

Agentic workflows

How to decide when a workflow should be agentic, deterministic, or a hybrid of both.

Read guide

AI agent frameworks

A practical checklist for comparing LangGraph, CrewAI, AutoGen-style systems, Eve, and other agent frameworks.

Read guide

AI agent tools

How to design agent tools with narrow authority, typed inputs, clear descriptions, and observable results.

Read guide

MCP vs skills

A grounded distinction between MCP servers, tools, resources, prompts, and skills for teams extending AI agents.

Read guide

Agent registry

What an AI agent registry must expose before developers can reuse agents safely.

Read guide

Shadcn registry for agents

How the shadcn registry model applies to AI agent workflows, where users need source files rather than opaque packages.

Read guide

LangGraph vs CrewAI

A builder-focused comparison of LangGraph and CrewAI, with Eve as a file-based reference point.

Read guide