Pioneering Autonomous Penetration Testing with Large Language Models through Prompt Engineering and Agentic System Design
Bibliographic record
Abstract
Autonomous Cyber Operations(ACO) aims to solve the ongoing cyber defense challenges caused by the prevalent cybersecurity talent shortage. Designing structured prompts with agentic systems can effectively direct Large Language Models(LLMs) behavior to navigate the attack through complex, multiphase operations without human oversight, leading to a fully autonomous cyber penetration testing and continuous cybersecurity posture monitoring. Current approaches to automated cyber-attacks lack the flexibility and adaptability to navigate the complex attack phases. Research in ACO has explored Artificial Intelligence(AI)-driven solutions, but integration of LLMs and prompt engineering strategies into these systems do not exist to date. This thesis introduces a novel phase-driven prompting methodology, called PromptPilot, paired with techniques such as Chain of Thought(CoT), Tree of Thought(ToT), and ReAct, to guide LLMs through the Cyber Kill Chain. Real-time trials in the simulated environment Emulated Cybernetic Hostile Operations(E.C.H.O) confirmed the viability of these prompt-driven autonomous penetration testing through exploitation. These results highlight that this emerging approach is viable, efficient, and precise in task execution across the attack phases, toward developing ACO agents. This research establishes the first framework for AI-driven autonomous penetration testing, emphasizing prompt and agentic system design as a cornerstone in advancing automated offensive and defensive cybersecurity capabilities.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".