Learning Map
This tutorial follows the order of real agent engineering work: build the smallest useful capability, then add requirements, context, tools, evaluation, and release gates to the same quality loop.
Quick StartCreate the first Skill and feedback loopSkillsPackage reusable capabilities and prompt assetsMCPExpose tools, resources, and context to agentsOpenSpecLock requirements and acceptance criteriaHarnessValidate output with scenarios, mocks, and evaluatorsEvaluationBuild regression suites and release gates
Recommended paths
Beginner path
- Quick Start - Run a minimal example first.
- What is Skill - Understand reusable capability boundaries.
- MCP Introduction - Learn how agents connect to external context.
- OpenSpec Core Concepts - Define intent before implementation.
- Harness Testing Infrastructure - Protect AI output quality with tests.
Engineering path
- Skill Practice - Package a task into a reusable capability.
- Build an MCP Server - Expose tools, resources, and safety boundaries.
- Complete Workflow - Move from proposal to implementation to archive.
- Evaluation and Quality - Turn acceptance criteria into regression checks.
- Deployment and Safety - Handle permissions, secrets, sandboxing, and rollback.
Architecture path
- Context Engineering - Design context windows, memory, and information priority.
- Workflow Orchestration - Connect Skills, MCP, OpenSpec, and Harnesses.
- Retrieval and Knowledge - Bring docs and codebase knowledge into the agent.
- Best Practices - Avoid unmaintainable prompt and tool design.
- Contributing - Extend the tutorial with the same quality bar.
Module standard
| Standard | Meaning |
|---|---|
| Concept | Explain the problem and vocabulary |
| Operation | Provide executable steps or minimal configuration |
| Practice | Include a task learners can self-check |
| Troubleshooting | Name common failure modes and diagnosis steps |
| Next step | Link to the next chapter or example |
Start now
Begin with Quick Start. If you already know the basics, jump to MCP Introduction or Evaluation and Quality.