Certified Agentic AI Pentester
(C-AgAIPen)
The Certified Agentic AI Pentester (C-AgAIPen) is an intermediate-level exam designed to test a candidate’s understanding of the fundamental concepts related to Agentic AI and GenAI application security. To pass this exam, candidates must demonstrate practical knowledge and the ability to conduct an Agentic AI application pentest, including identifying vulnerabilities aligned with the OWASP Agentic AI Top 10.
Note: The exam details will be sent to you on/before 20th April, 2026.
- Practical
- 4 Hours
- Online
- On-demand
- Real world pentesting scenarios
£250.00
£50.00
Who should take the exam?
Certified Agentic AI Pentester (C-AgAIPen) is intended to be taken by pentesters, application security architects, SOC analysts, red and blue team members, AI/ML engineers, AI product security professionals, and any AI security enthusiast who wants to evaluate and advance their knowledge in securing Agentic AI systems.
What is the format of the exam?
Certified Agentic AI Pentester (C-AgAIPen) is an intense 4-hour practical exam. It requires candidates to solve a number of challenges, identify and exploit various vulnerabilities, and obtain flags. The exam can be taken online, anytime (on-demand), and from anywhere. Candidates will need to connect to the exam VPN server to access the vulnerable Agentic AI application.
What is the pass criteria for the exam?
The pass criteria are as follows:
- Candidates scoring over 60% marks will be deemed to have successfully passed the exam.
- Candidates scoring over 75% marks will be deemed to have passed with merit.
What is the experience needed to take the exam?
This is an intermediate-level exam. Candidates should have prior knowledge and experience in AI/ML and Agentic AI application pentesting. They should have a solid understanding of common application security topics, including the OWASP Top 10 for LLM applications, prompt injection attacks, insecure output handling, and common security misconfigurations, as well as the OWASP Top 10 for Agentic Applications 2026: Agent Goal Hijack, Tool Misuse & Exploitation, Identity & Privilege Abuse, Agentic Supply Chain Vulnerabilities, Unexpected Code Execution (RCE), Memory & Context Poisoning, Insecure Inter-Agent Communication, Cascading Failures, Human-Agent Trust Exploitation, and Rogue Agents.
Candidates should be able to demonstrate practical knowledge by identifying and exploiting vulnerabilities in real-world Agentic AI scenarios.
Note: As this is an intermediate-level exam, a minimum of two years of professional pentesting experience is recommended.
What will the candidates get?
On completing the exam, each candidate will receive:
- A certificate with their pass/fail and merit status.
- The certificate will contain a certificate number, which can be used by anyone to validate the certificate.
What is the exam retake policy?
Candidates, who fail the exam, are allowed 1 free exam retake within the exam fees.
What are the benefits of this exam?
The exam allows candidates to demonstrate their expertise in Agentic AI security and application pentesting, including assessing multi-agent systems, tool integrations, and autonomous decision-making workflows. Achieving this certification validates practical, hands-on skills aligned with modern Agentic AI threats and helps professionals advance their careers in AI security, red teaming, and AI governance roles.
How long is the certificate valid for?
The certification does not have an expiration date. However, the passing certificate will mention the details of the exam, such as the exam version and the date. As the exam is updated over time, candidates should retake the newer version as per their convenience.
Are candidates allowed to use AI tools during the exam?
No, candidates are not allowed to use AI in any way during the exam.
What is the refund policy?
All purchases are final. We do not offer refunds once a transaction has been completed.
What is the difference between C-AgAIPen and C-AI/MLPen?
C-AgAIPen:
The Certified Agentic AI Pentester exam focuses on security testing of autonomous AI agents that can plan, reason, and execute multi-step actions. It emphasizes vulnerabilities in agent workflows, tool integrations, memory handling, and decision-making logic. Candidates are expected to assess risks such as task manipulation, tool abuse, goal hijacking, and chained prompt attacks across agent pipelines, simulating real-world agentic system exploitation.
C-AI/MLPen:
The Certified AI/ML Pentester exam focuses on identifying and exploiting vulnerabilities in machine learning models and large language model (LLM) applications. It covers areas such as prompt injection, data leakage, model misconfigurations, and OWASP Top 10 for LLMs. The exam is practical and scenario-based, requiring candidates to interact with AI systems, analyze their behavior, and extract sensitive information or bypass security controls in a controlled lab environment.
Note: It is recommended to attempt the Certified AI/ML Pentester exam first to build a strong foundation in AI security concepts before progressing to the more advanced Certified Agentic AI Pentester exam.
Will you provide any training that can be taken before the exam?
Being an independent certifying authority, we (The SecOps Group) do not provide any training for the exam. Candidates should carefully go over each topic listed in the syllabus and make sure they have adequate understanding, required experience and practical knowledge of these topics. Additionally, the following independent resources can be used to prepare for the exam.
Learning Resources
Exam Syllabus
Agent Goal Hijack
- Direct Prompt Injections
- Indirect Prompt Injections
Sensitive Information Disclosure
- Exposure of system prompts
- Leakage of API keys, credentials, or internal data
- Data exfiltration through prompts or agent tools
Insecure Output Handling
- Execution of unsafe AI-generated outputs
- Command or script execution through model responses
Training Data Poisoning
- Manipulation of training datasets
- Injection of malicious content into model knowledge sources
Model Denial of Service
- Prompt techniques causing excessive resource consumption
- Token exhaustion or model overload attacks
Supply Chain Vulnerabilities
- Risks from third-party models, plugins, or integrations
- Compromise of external AI dependencies
Tool Misuse and Exploitation
Identity and Privilege Abuse
- Exploitation of weak authentication mechanisms in agent workflows
- Privilege escalation through misconfigured roles or excessive permissions
Agentic Supply Chain Vulnerabilities
RAG Pipelines Exploitation
Model Context Protocol (MCP) Tool Exploitation
Unexpected Code Execution (RCE)
Memory & Context Poisoning
- Manipulation of contextual history to alter agent decisions
- Contamination of shared memory between agents
Insecure Inter-Agent Communication
Cascading Failures
- Failure of one agent triggering downstream agent malfunctions
- Propagation of errors through multi-agent workflows