ASECCMTMaster ClassMT-ASECC1.0<p>The exercises will be conducted in Python, TypeScript, and C#. Basic knowledge of these programming languages is helpful but not required.</p><p>Software Architects, Developer (All)</p><h5>Module 1: Claude Code Fundamentals</h5><ul> <li>Introduction to Claude Code</li><li>Setup &amp; Installation</li><li>Editors: Console, Desktop App, VS Code</li><li>CLAUDE.md constitution, @-mentioning files and URLs</li><li>Prompt caching, cost and token awareness</li><li>Model Selection Strategy</li><li>Permission Model: Tools, Bash, Network</li><li>Git worktrees and the agents view for parallel sessions</li><li>Artifacts: results as shareable pages</li></ul><h5>Module 2: Claude Code Configure Harness</h5><ul> <li>Settings Hierarchy: Global, Project, Session</li><li>Rules and context via CLAUDE.md</li><li>Slash Commands</li><li>Hooks: PreToolUse, PostToolUse, Notification, Stop</li><li>MCP Servers &amp; Tools</li><li>Skill folders and their sources (module 3 owns authoring)</li><li>Plugins as distribution wrapper (module 11 owns marketplaces)</li><li>Auto-memory: MEMORY.md index, four typed files</li><li>Checkpoints &amp; Session Safety</li><li>Sub-agents as a harness file type (module 7 owns orchestration)</li></ul><h5>Module 3: Creating &amp; Managing Skills</h5><ul> <li>Skills vs. slash commands</li><li>Invocation model, anatomy, body structure</li><li>Authoring: manual, or extracted from a conversation</li><li>Front matter fields</li><li>Allowed-tools restrictions and metadata conventions</li><li>Composite skills: scripts, references, assets</li><li>Cross-skill references and delegation patterns</li><li>Dynamic context injection: command preprocessing</li><li>Forked execution: context fork, agent selection, background</li><li>Testing: Trigger, correctness, performance</li><li>Distribution: portable folder, skills API, hosting</li></ul><h5>Module 4: Consuming &amp; Designing MCP Tools</h5><ul> <li>MCP architecture: servers, clients, transports</li><li>Session lifecycle: what servers and clients expose</li><li>Remote auth: bearer tokens, API keys, OAuth 2.0</li><li>Building a server in C#: Streamable HTTP, EF Core</li><li>Inspection with the MCP Inspector</li><li>MCP Apps: interfaces rendered inline in the host</li><li>Artifact connectors: pages calling MCP at view time</li><li>Tool design: granularity, descriptions, resources</li><li>Error and deferred-loading contract</li></ul><h5>Module 5: Agentic CLI &amp; Headless Execution</h5><ul> <li>Agent harness: tools, knowledge, action interfaces</li><li>CLI, terminal, and where headless mode fits</li><li>Headless mode: print, JSON output, CI exit codes</li><li>Least-privilege tool scoping</li><li>CI gates: sentinel tokens, deterministic exit codes</li><li>GitHub Actions: scripting agentic workflows</li><li>Parallel worktree lanes and task-queue orchestration</li><li>PreToolUse policy gates across lanes</li><li>Always-on goal patterns with turn limits</li></ul><h5>Module 6: Planning &amp; Spec-Driven Development</h5><ul> <li>Plan Mode: mechanics and activation</li><li>Stepwise context: questions and constraints</li><li>Spec iteration via feedback loops</li><li>Extended Thinking: steering reasoning chains</li><li>Thinking budget vs. token cost</li><li>Authoring specs and technical plans</li><li>Decomposing tasks into sub-agent packages</li><li>Defining and validating done criteria</li><li>Spec Kit and BMAD-METHOD against hand-driven Plan Mode</li></ul><h5>Module 7: Sub-Agents</h5><ul> <li>Where single-agent sequential hits a ceiling</li><li>Frontmatter: name, model, tools, permission mode</li><li>Restricting tools and permissions per agent</li><li>Backend agent: .NET, conventions skill, Microsoft Learn MCP</li><li>Frontend agent: Angular, Angular CLI and DevTools MCPs</li><li>QA agent: Playwright, no implementation authority</li><li>Evaluating performance: test cases, edge cases, failures</li><li>Orchestration: prompt delegation, loops, workflow graphs</li><li>Context isolation between parallel agents</li></ul><h5>Module 8: Agent Teams</h5><ul> <li>Where a team beats a sub-agent, and where it loses</li><li>Lead, teammates, shared task list, mailbox</li><li>Enabling teams and what lands on disk</li><li>Spawning teammates: names and models</li><li>Display modes, agent panel, steering keys</li><li>Task states, dependencies, file-locked claiming</li><li>Messaging: lead to teammate, teammate to teammate</li><li>Competing hypotheses: teammates disproving each other</li><li>Plan approval and the team lifecycle hooks</li><li>Team sizing, file conflicts, context window cost</li></ul><h5>Module 9: Optimizing the Agentic Harness &amp; Continuous Learning</h5><ul> <li>Skills and agents as living artifacts</li><li>Autoresearch loop: propose, evaluate, revert</li><li>Deterministic experiments for skill quality</li><li>Caveman skill: 60-75% output compression</li><li>Compressing CLAUDE.md with caveman-mode</li><li>Memory architectures: in-context loading, statefile, knowledge graph</li><li>Updating harness from observed behavior</li><li>Metrics: determinism, output structure, abort rate</li></ul><h5>Module 10: Security, Hardening &amp; Observability</h5><ul> <li>Prompt injection: tool results and files</li><li>Tool poisoning: validating MCP responses</li><li>Untrusted content isolation in agent context</li><li>Guardrails: hard stops and scope constraints</li><li>Least privilege scoping per sub-agent</li><li>Secrets: preventing cross-agent credential leakage</li><li>Tracing tool chains and cost attribution</li><li>Diagnosing stuck, runaway, hallucinating agents</li><li>Artifact governance: toggles, retention, audit events</li></ul><h5>Module 11: Bundling &amp; Distributing Plugins</h5><ul> <li>Plugins vs. standalone configuration: namespacing</li><li>Plugin anatomy: manifest, skills, hooks, MCP</li><li>Bundling skills, hooks, MCP configs</li><li>LSP servers and background monitors</li><li>Default agent activation via plugin settings</li><li>Local testing: plugin directory and reload</li><li>Converting a .claude configuration into a plugin</li><li>Publishing to official and team marketplaces</li></ul><h5>Module 12: Agentic DevOps &amp; Deployment</h5><ul> <li>The deployment sub-agent: allowlists, prove-it contract</li><li>Bootstrapping and hardening Ubuntu over SSH</li><li>Docker Compose behind a Caddy edge, automatic TLS</li><li>DNS and ACME ordering: the trap retrying cannot fix</li><li>Core Web Vitals and agentic SEO loops</li></ul><h5>Module 13: Claude Managed Agents</h5><ul> <li>Cloud containers with network and files</li><li>Steering agents mid-execution via SSE</li><li>OAuth vaults and token refresh</li><li>Callable-agent patterns with isolated threads</li><li>Session lifecycle: idle, running, termination</li><li>Human-in-the-loop approval gates</li><li>Managed Agents vs. Messages API</li><li>Versioning agents as deployable units</li></ul><h5>Module 14: Claude Code SDK</h5><ul> <li>Spawning Claude Code via the SDK</li><li>Sessions, context, streaming responses</li><li>Custom tools as in-process MCP servers</li><li>Typed output with Zod, SDK in CI</li><li>Parallel agents: session isolation, orchestrator pattern</li><li>Output control: tool choice, Structured Outputs, Batches</li></ul>The exercises will be conducted in Python, TypeScript, and C#. Basic knowledge of these programming languages is helpful but not required.Software Architects, Developer (All)Module 1: Claude Code Fundamentals - Introduction to Claude Code - Setup & Installation - Editors: Console, Desktop App, VS Code - CLAUDE.md constitution, @-mentioning files and URLs - Prompt caching, cost and token awareness - Model Selection Strategy - Permission Model: Tools, Bash, Network - Git worktrees and the agents view for parallel sessions - Artifacts: results as shareable pages Module 2: Claude Code Configure Harness - Settings Hierarchy: Global, Project, Session - Rules and context via CLAUDE.md - Slash Commands - Hooks: PreToolUse, PostToolUse, Notification, Stop - MCP Servers & Tools - Skill folders and their sources (module 3 owns authoring) - Plugins as distribution wrapper (module 11 owns marketplaces) - Auto-memory: MEMORY.md index, four typed files - Checkpoints & Session Safety - Sub-agents as a harness file type (module 7 owns orchestration) Module 3: Creating & Managing Skills - Skills vs. slash commands - Invocation model, anatomy, body structure - Authoring: manual, or extracted from a conversation - Front matter fields - Allowed-tools restrictions and metadata conventions - Composite skills: scripts, references, assets - Cross-skill references and delegation patterns - Dynamic context injection: command preprocessing - Forked execution: context fork, agent selection, background - Testing: Trigger, correctness, performance - Distribution: portable folder, skills API, hosting Module 4: Consuming & Designing MCP Tools - MCP architecture: servers, clients, transports - Session lifecycle: what servers and clients expose - Remote auth: bearer tokens, API keys, OAuth 2.0 - Building a server in C#: Streamable HTTP, EF Core - Inspection with the MCP Inspector - MCP Apps: interfaces rendered inline in the host - Artifact connectors: pages calling MCP at view time - Tool design: granularity, descriptions, resources - Error and deferred-loading contract Module 5: Agentic CLI & Headless Execution - Agent harness: tools, knowledge, action interfaces - CLI, terminal, and where headless mode fits - Headless mode: print, JSON output, CI exit codes - Least-privilege tool scoping - CI gates: sentinel tokens, deterministic exit codes - GitHub Actions: scripting agentic workflows - Parallel worktree lanes and task-queue orchestration - PreToolUse policy gates across lanes - Always-on goal patterns with turn limits Module 6: Planning & Spec-Driven Development - Plan Mode: mechanics and activation - Stepwise context: questions and constraints - Spec iteration via feedback loops - Extended Thinking: steering reasoning chains - Thinking budget vs. token cost - Authoring specs and technical plans - Decomposing tasks into sub-agent packages - Defining and validating done criteria - Spec Kit and BMAD-METHOD against hand-driven Plan Mode Module 7: Sub-Agents - Where single-agent sequential hits a ceiling - Frontmatter: name, model, tools, permission mode - Restricting tools and permissions per agent - Backend agent: .NET, conventions skill, Microsoft Learn MCP - Frontend agent: Angular, Angular CLI and DevTools MCPs - QA agent: Playwright, no implementation authority - Evaluating performance: test cases, edge cases, failures - Orchestration: prompt delegation, loops, workflow graphs - Context isolation between parallel agents Module 8: Agent Teams - Where a team beats a sub-agent, and where it loses - Lead, teammates, shared task list, mailbox - Enabling teams and what lands on disk - Spawning teammates: names and models - Display modes, agent panel, steering keys - Task states, dependencies, file-locked claiming - Messaging: lead to teammate, teammate to teammate - Competing hypotheses: teammates disproving each other - Plan approval and the team lifecycle hooks - Team sizing, file conflicts, context window cost Module 9: Optimizing the Agentic Harness & Continuous Learning - Skills and agents as living artifacts - Autoresearch loop: propose, evaluate, revert - Deterministic experiments for skill quality - Caveman skill: 60-75% output compression - Compressing CLAUDE.md with caveman-mode - Memory architectures: in-context loading, statefile, knowledge graph - Updating harness from observed behavior - Metrics: determinism, output structure, abort rate Module 10: Security, Hardening & Observability - Prompt injection: tool results and files - Tool poisoning: validating MCP responses - Untrusted content isolation in agent context - Guardrails: hard stops and scope constraints - Least privilege scoping per sub-agent - Secrets: preventing cross-agent credential leakage - Tracing tool chains and cost attribution - Diagnosing stuck, runaway, hallucinating agents - Artifact governance: toggles, retention, audit events Module 11: Bundling & Distributing Plugins - Plugins vs. standalone configuration: namespacing - Plugin anatomy: manifest, skills, hooks, MCP - Bundling skills, hooks, MCP configs - LSP servers and background monitors - Default agent activation via plugin settings - Local testing: plugin directory and reload - Converting a .claude configuration into a plugin - Publishing to official and team marketplaces Module 12: Agentic DevOps & Deployment - The deployment sub-agent: allowlists, prove-it contract - Bootstrapping and hardening Ubuntu over SSH - Docker Compose behind a Caddy edge, automatic TLS - DNS and ACME ordering: the trap retrying cannot fix - Core Web Vitals and agentic SEO loops Module 13: Claude Managed Agents - Cloud containers with network and files - Steering agents mid-execution via SSE - OAuth vaults and token refresh - Callable-agent patterns with isolated threads - Session lifecycle: idle, running, termination - Human-in-the-loop approval gates - Managed Agents vs. Messages API - Versioning agents as deployable units Module 14: Claude Code SDK - Spawning Claude Code via the SDK - Sessions, context, streaming responses - Custom tools as in-process MCP servers - Typed output with Zod, SDK in CI - Parallel agents: session isolation, orchestrator pattern - Output control: tool choice, Structured Outputs, Batches4 days4890.004890.004890.004890.004890.004890.004180.004890.00