AgenticSwift

The course

The first module is free to read. Unlock the course to open the rest and save your place.

Unlock the course

  1. 1Agent Harness Foundations

    Free

    Why fast progress with AI stalls, what an agent harness is and is not, one turn of the loop, the eight dimensions of an iOS harness and a mission followed from start to delivery.

    1. The 70% problem
    2. Agent, harness, harness engineering
    3. One turn of the agent loop
    4. Eight dimensions of an iOS agent harness
    5. A mission from start to delivery
    6. Exercise, baseline versus configured

  2. 2Building Agent-Ready Swift Repositories

    Make a Swift codebase easy for an agent to explore, build and change without guessing.

    1. Repository structure and discoverability
    2. Modular Swift architecture with Swift Package Manager
    3. AGENTS.md and CLAUDE.md
    4. Architecture documentation and Xcode exploration
    5. Exercise, prepare AgenticShop for agents

  3. 3Context Engineering for Swift Developers

    Decide what an agent reads, when it reads it, and what it should never have to carry.

    1. Context windows and saturation
    2. Repository exploration and retrieval
    3. Persistent memory, ADRs and feature specifications
    4. Mission-specific context and isolation
    5. Exercise, two context strategies for one feature

  4. 4Extending Agents with Tools, Skills and MCP

    Give an agent capabilities it does not have, described so that it uses them at the right moment.

    1. Tools and tool calling
    2. Agent Skills and reusable workflows
    3. The Model Context Protocol
    4. Building Swift developer utilities
    5. Exercise, swiftProjectAnalyzer over MCP

  5. 5Feedback Loops and Automated Validation

    Let the compiler and the tests answer the agent directly, so that nobody has to relay an error message.

    1. Agent feedback loops
    2. Compiler diagnostics and xcodebuild
    3. Swift Testing, XCTest and test-driven agentic development
    4. Quality gates and Definition of Done
    5. Exercise, a self-correcting Shopping Cart workflow

  6. 6Agent Governance and Security

    Decide what an agent may do, what needs a person, and what it must never reach.

    1. Permission models and approvals
    2. Secrets and sandboxes
    3. Prompt injection and untrusted instructions
    4. Audit logs and cost limits
    5. Exercise, secure an agent environment

  7. 7Multi-Agent Development and Git Worktrees

    Run several agents on one repository, keep their work apart, and make the pieces fit when they meet.

    1. Single-agent versus multi-agent development
    2. Missions, boundaries and contracts
    3. Git worktrees and parallel execution
    4. Roles and integration
    5. Exercise, shopping Cart with several agents

  8. 8Evaluation and Production Readiness

    Replace impressions with measurements, then decide whether a harness is ready for a team.

    1. What to measure
    2. Scenarios and execution traces
    3. Token consumption and cost management
    4. CI/CD integration and production readiness
    5. Exercise, evaluate harness configurations