Course Outline
From autocomplete to agents: understanding agent failures
• Anatomy of a coding agent: model, harness, tool surface, context, permissions.
• Positioning each tool: Claude Code, GitHub Copilot, Cursor, Codex CLI, and Gemini CLI.
• A taxonomy of failure causes: incorrect context, inappropriate tools, lack of feedback, and unbounded autonomy.
Demonstration: Executing the same task successfully and unsuccessfully side by side.
Context engineering
• Treating the context window as a budget: determining what earns a spot within it.
• AGENTS.md, CLAUDE.md, .cursor/rules, copilot-instructions.md: one concept applied across multiple filenames to establish a single source of truth.
• Establishing conventions, build and test commands, and architectural boundaries.
• Differentiating retrieval from explicit context; managing task decomposition and sub-agents.
Lab: Creating repository context for an unfamiliar Python service, then re-running a failed task to compare outputs.
Reusable workflows and Agent Skills
• Selecting the right abstraction: instruction file, skill, custom command, or plain script.
• Deconstructing a skill: triggering mechanisms, instructions, bundled scripts, and progressive disclosure.
• Ensuring portability across tools and identifying where vendor lock-in begins.
• Versioning, reviewing, and distributing skills across a team; recognizing common anti-patterns.
Lab: Designing and testing a reusable workflow that enforces internal coding standards.
MCP: Linking agents to operational systems
• Architectural components: clients, servers, tools, resources, and prompts; utilizing stdio and HTTP transports.
• Identifying valuable servers: Git hosting platforms, issue trackers, databases, browsers, and internal APIs.
• Determining when a CLI or script is more effective than an MCP server.
• Maintaining tool-surface hygiene: why an excessive number of tools reduces reliability.
Lab: Connecting MCP servers and managing a ticket end-to-end—issue, branch, patch, tests, and pull request.
Feedback loops and evaluation
• Using tests, types, and linters as the agent's ground truth; employing test-first approaches as a control mechanism.
• Leveraging CI as an outer loop and maintaining review discipline for agent-authored diffs.
• Creating golden-task evaluation sets: defining what to measure and detecting regressions.
• Treating cost and latency as primary metrics.
Lab: Constructing a small evaluation set and scoring two different agent configurations against it.
Security and guardrails
• Mitigating prompt injection via issues, pull requests, READMEs, dependencies, and fetched web pages.
• Implementing permission models: allowlists, approval workflows, read-only tools, and network egress controls.
• Ensuring secret hygiene and sandboxing: using containers, ephemeral credentials, and limiting blast radius.
• Addressing supply-chain risks in third-party MCP servers and shared skills.
Lab: Observing an agent being compromised by a poisoned repository, then hardening the environment to prevent recurrence.
Deploying this approach within a team
• Following a staged adoption path; deciding what to standardize and what to leave to individual preference.
• Identifying metrics that demonstrate genuine value versus those that do not.
Requirements
• Practical familiarity with Python, Git, and the command line
• Prior experience using an AI coding assistant
NobleProg will provision Dadesktop VMs for participants, pre-installed with Docker, VS Code, and Python 3.11 or newer.
• A functioning AI coding assistant of the participant's preference: Claude Code, GitHub Copilot, Cursor, Codex CLI, or Gemini CLI. The labs are tool-agnostic, and instructions are supplied for each option.
Audience
• Software engineers, tech leads, and architects who struggle to achieve reliable results with AI coding assistants.
• Platform and developer-experience engineers deploying AI tools across multiple teams.
• Engineering managers establishing standards, guardrails, and success metrics.
Testimonials (3)
The trainer is patient and very helpful. He knows the topic well.
CLIFFORD TABARES - Universal Leaf Philippines, Inc.
Course - Agentic AI for Business Automation: Use Cases & Integration
Good mixvof knowledge and practice
Ion Mironescu - Facultatea S.A.I.A.P.M.
Course - Agentic AI for Enterprise Applications
The mix of theory and practice and of high level and low level perspectives
