Training course · AI Agent Development · Security & DevSecOps

AI Agent Development for Security and DevSecOps Teams

Build AI agents professionally: architectures, tool use, agent SDKs, LangGraph, memory, evals, guardrails and deployment.

Duration
2 days (14 hrs)
Format
Online or on-site
Language
German or English
Participants
open dates or exclusive for your team
Certificate
Certificate of attendance

Overview

“AI Agent Development for Security and DevSecOps Teams” is designed for roles such as Application security engineers and security champions in development teams, DevSecOps and platform engineers responsible for CI/CD pipelines and Penetration testers and red team members and shows step by step how to use AI Agent Development productively, safely and transparently in Security & DevSecOps – from the fundamentals to advanced workflows.

AI agents go beyond conventional chatbots: they break tasks down, call tools and APIs, evaluate the results and keep iterating until a goal is met. In this two-day course you learn to build such systems so that they work reliably, transparently and safely in day-to-day business. You compare architectural patterns ranging from simple workflows through the agent loop to multi-agent systems, and you implement them hands-on with the Claude Agent SDK, the OpenAI Agents SDK and LangGraph.

The focus is not on a quick demo but on the path to production. You design tool interfaces and context deliberately, connect tools via the Model Context Protocol, plan short-term and long-term memory and build evaluations that surface regressions early. On top of that come guardrails against prompt injection and harmful actions, human-in-the-loop approvals, tracing, cost control and deployment. By the end you can judge when an agent is worthwhile, when a deterministic workflow is enough and which framework fits your project.

AI coding assistants and agents affect application security from two directions. They speed up threat modelling, code review and detection engineering, yet they also produce new code faster than traditional security gates can absorb. Security and DevSecOps teams therefore use AI to triage SAST findings, propose fixes for vulnerabilities, harden CI/CD pipelines and scan infrastructure as code for misconfigurations. What matters is a controlled setup: agents running with least privilege, no secrets in the prompt context, every change traceable through a pull request and human sign-off on security-critical decisions. In this course you learn to use AI as a force multiplier for your security programme while keeping the new risks of generated code, prompt injection and insecure tool integrations firmly under control.

The course is delivered live – either online in a virtual classroom or on-site in our training rooms in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart and Kaiserslautern. We can also run it exclusively for your team.

Who is this course for?

  • Application security engineers and security champions in development teams
  • DevSecOps and platform engineers responsible for CI/CD pipelines
  • Penetration testers and red team members
  • SOC analysts and detection engineers
  • Developers accountable for secure software supply chains

Typical challenges we address

  • A flood of SAST, DAST and SCA findings with a high share of false positives
  • Security review becomes a bottleneck as AI-assisted teams ship far more code
  • Misconfigurations in Terraform, Kubernetes manifests and cloud policies go unnoticed
  • New attack surfaces from LLM integrations, prompt injection and over-privileged agents

What you will be able to do

  • You choose the right pattern for a task, from a fixed workflow to an autonomous agent.
  • You design tools and context so that agents work reliably and cost-effectively.
  • You implement agents with the Claude Agent SDK, the OpenAI Agents SDK or LangGraph.
  • You set up evaluations and tracing to measure quality and detect regressions.
  • Use AI agents for triage, review and fix generation without giving up control over security-critical changes.
  • Integrate AI-assisted checks into CI/CD pipelines alongside your existing SAST, SCA and IaC scanners.
  • Assess and mitigate the specific risks of AI-generated code, prompt injection and agent tool access.

Course outline

Day 1

Module 1 Agent architectures and design patterns
  • Workflows versus agents: prompt chaining, routing, parallelisation, orchestrator-worker
  • The agent loop: plan, act, observe, reflect
  • Multi-agent systems, subagents and handoffs between agents
  • Criteria for choosing model, framework and level of autonomy
Module 2 Tool use and context engineering
  • Function calling and structured outputs with JSON Schema
  • Designing good tools: granularity, naming, error messages, idempotency
  • Connecting and reusing tools via MCP servers
  • Managing the context window: summarisation, compaction, retrieval and prompt caching
Module 3 Frameworks in practice
  • Claude Agent SDK: agent loop, tools, subagents, permissions and hooks
  • OpenAI Agents SDK: agents, handoffs, guardrails and tracing
  • LangGraph: stateful graphs, checkpointing and human-in-the-loop
  • Side-by-side comparison on the same task and criteria for choosing a framework
Module 4 Memory, evals and observability
  • Short-term and long-term memory, storage strategies and forgetting
  • Evaluation datasets, LLM-as-a-judge and task-based success metrics
  • Tracing tool calls, token usage and latency
  • Regression testing when prompts, models or tools change

Day 2

Module 5 Guardrails, security and deployment
  • Protecting against prompt injection, data exfiltration and over-privileged tools
  • Input and output checks, approval steps and sandboxing
  • Deploying as a service, asynchronous execution, retries and cost limits
  • Governance, audit logging and requirements under GDPR and the EU AI Act
Module 6 Practice lab: Security & DevSecOps
  • Faster threat modelling
  • SAST finding triage
  • Secure code review
  • Hardening IaC and containers
  • Detection engineering
  • Securing the supply chain
Module 7 Data protection, quality & transfer to daily work
  • Source code, vulnerability data and logs are highly sensitive: you learn which data may be shared with AI services, how to keep secrets out of the context and how AI use aligns with NIS2, the Cyber Resilience Act, the GDPR and the EU AI Act.
  • Ground rules for safe AI use in your team
  • Checking results: facts, sources, tests and the four-eyes principle
  • Personal action plan for the next 30 days
  • Open questions, outlook and resources for further learning

Practical scenarios from your field

Faster threat modelling

Derive STRIDE-based threat analyses from architecture descriptions and data flow diagrams, identify trust boundaries and turn countermeasures into testable requirements ready for the backlog. The model is updated rather than rebuilt when the architecture changes.

SAST finding triage

Match static analysis results against real code context, assess reachability and exploitability, dismiss false positives with a documented rationale and prepare minimal fixes for genuine vulnerabilities as pull requests.

Secure code review

Have pull requests checked specifically for injection flaws, broken authorisation, insecure deserialisation and cryptographic mistakes, with references to OWASP categories and a clear explanation for every finding.

Hardening IaC and containers

Analyse Terraform modules, Helm charts and Dockerfiles for over-privileged roles, exposed ports, missing encryption and unsafe base images, then generate hardened versions backed by policy-as-code rules.

Detection engineering

Draft Sigma rules, YARA signatures and SIEM queries from threat descriptions, validate them against test data and enrich documentation with the relevant MITRE ATT&CK techniques.

Securing the supply chain

Evaluate SBOMs, prioritise vulnerable dependencies, analyse upgrade paths including breaking changes and add pipeline steps for signing, provenance attestation and dependency pinning. Each update pull request explains the risk.

Possible hands-on projects

Finding triage bot

A script that ingests SARIF output from the pipeline, deduplicates findings, enriches them with code context and opens prioritised tickets with a suggested fix.

Secret scanning hook

A pre-commit hook with custom patterns for internal token formats, allowlist management and clear guidance on how to rotate a leaked secret.

Threat model generator

A small web app that turns a system description into a structured threat model with data flows, risks and mitigations, exportable as a document.

Prerequisites

Solid programming skills in Python or TypeScript, experience with REST APIs and some hands-on experience with LLM APIs or AI assistants. You have hands-on experience in software development or operations and a working knowledge of application security, such as the OWASP Top 10 and CI/CD pipelines.

Technical requirements

A laptop with Python or Node.js, Git and a code editor, plus API access to at least one LLM provider such as Anthropic or OpenAI. Credentials for the exercises can be provided if needed.

Data protection & compliance

Source code, vulnerability data and logs are highly sensitive: you learn which data may be shared with AI services, how to keep secrets out of the context and how AI use aligns with NIS2, the Cyber Resilience Act, the GDPR and the EU AI Act.

Suggested dates

All dates are suggestions – your preferred date is always possible.

  • Tue, 24 November 2026 – Wed, 25 November 2026 Online Request
  • Thu, 3 December 2026 – Fri, 4 December 2026 Munich Request
  • Wed, 9 December 2026 – Thu, 10 December 2026 Online Request
  • Tue, 15 December 2026 – Wed, 16 December 2026 Stuttgart Request
  • Mon, 11 January 2027 – Tue, 12 January 2027 Online Request
  • Wed, 20 January 2027 – Thu, 21 January 2027 Berlin Request

Frequently asked questions

Which agent framework should we use?

It depends on your model strategy, the level of control you need and your team’s skills. Vendor SDKs offer a lean start with tightly integrated features, while LangGraph gives you finer control over state and flow. In the course you solve the same task with several frameworks and derive clear selection criteria for your own project.

How do you stop an agent from taking wrong or dangerous actions?

With several layers of protection: narrowly scoped tools with minimal permissions, input and output validation, approval steps for critical actions, sandboxing and continuous evaluation. Prompt injection via external content is treated as a realistic threat. You practise these measures directly on your own agent.

Does every automation need an agent?

No. Many tasks are handled more reliably, cheaply and transparently by a fixed workflow of a few LLM calls. Agents pay off when the solution path cannot be planned in advance and the system must react flexibly to intermediate results. We train exactly this trade-off using practical examples.

Is AI-generated code a security risk?

AI assistants sometimes reproduce insecure patterns, outdated libraries or even non-existent package names. The risk is manageable when generated code passes the same gates as hand-written code: peer review, SAST, dependency scanning and tests. In the course you set up exactly these guardrails while using AI yourself to find and fix vulnerabilities faster.

Should an AI agent be allowed to apply security fixes on its own?

We recommend a tiered model: the agent analyses, proposes fixes and opens pull requests, while humans approve the merge. For low-risk routine work such as dependency updates with passing tests, more automation can make sense. You learn to design permissions, sandboxing and audit trails so every decision remains traceable.

How much does “AI Agent Development for Security and DevSecOps Teams” cost?

The course costs a flat €949 plus VAT per participant – whether online or on-site. This includes 2 training day(s) with a live trainer, course materials, a lab environment and a certificate of attendance.

Can I suggest my own date?

Yes. Our suggested dates are just a guide. In the request form you can always describe your preferred date or period in free text – we check availability and confirm within one working day.

Is the course online or on-site?

Both are possible: live online in a virtual classroom or on-site in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart or Kaiserslautern. We can also run the course exclusively for your team.

More courses for Security and DevSecOps Teams

More AI Agent Development courses

All 30 AI Agent Development courses

AI Agent Development · MunichAI Agent Development · SalzburgAI Agent Development · BerlinAI Agent Development · AugsburgAI Agent Development · HanoverAI Agent Development · BremenAI Agent Development · StuttgartAI Agent Development · KaiserslauternAI Agent Development · Online