With the advancement of the information age, the widespread application of electronic evidence in fields such as justice and finance has brought new challenges. Although existing blockchain electronic evidence sharing schemes have immutability and transparency, they still have shortcomings in access control, data privacy protection, and efficiency. In addition, traditional attribute encryption strategies lack effective revocation mechanisms and cannot fully protect privacy when implementing fine-grained access control. Therefore, in order to address the above limitations, a blockchain electronic evidence sharing scheme based on an improved ciphertext policy attribute encryption combined with zero knowledge proof technology has been proposed. The research innovatively introduces revocable ciphertext strategy encryption, which addresses the security risks caused by decryption key leakage through revocation function, ensuring the secure storage and sharing of electronic evidence. Meanwhile, the study also improved the PBFT consensus algorithm to enhance its performance in handling large volumes of transactions. The results showed that the storage TPS of the research model reached 492, and the query TPS reached 655. The computational cost of improving the PBFT consensus algorithm is 1.94 Ă 10 4 , and the maximum computational cost of the electronic evidence access control model based on zero knowledge proof is 509. Compared with traditional blockchain based electronic evidence sharing methods, the improved method not only enhances storage and sharing efficiency, but also further strengthens privacy protection capabilities by combining zero knowledge proof technology. In summary, the research method effectively achieves secure sharing and privacy protection of electronic evidence on blockchain, providing support and reference for electronic evidence storage in fields such as justice and finance. However, there are still challenges in terms of scalability and data storage in the research, so algorithms can be optimized in the future to further improve the application scope of the system.
Yuanfeng Xie, Weiwei Jiang, Hanqing Luo, Jian Ping Gan
This study presents an innovative authentication scheme that integrates Physical Unclonable Functions (PUFs) and Zero-Knowledge Proofs (ZKP) to provide efficient and secure authentication for Internet of Things (IoT) devices. Traditional PUF-based protocols offer strong security but incur high resource costs and slow authentication. To address this, we propose a joint scheme. First, a unified architecture combining a PUFâTrue Random Number Generator (TRNG) is introduced. This architecture utilizes a feedback permutation obfuscation mechanism and an arbitration delay deviation with a metastable design from a ring oscillator, ensuring the PUFâTRNG system possesses both attack resistance and true random properties. The architecture provides synchronization for both PUF and TRNG in the protocol. Next, we integrate Schnorrâs ZKP with a PUF-based key encapsulation and reconstruction scheme to construct an end-to-end anonymous identity authentication protocol that does not require real-time participation of a trusted third party. The protocol requires only two handshakes, significantly reducing the number of protocol rounds compared to related protocols. Finally, the PUFâTRNG architecture has been implemented on the Xilinx XC7A100T development board. Experimental results show that the PUF circuit effectively resists various modeling attacks. Formal verification with ProVerif demonstrates confidentiality, mutual authentication, and robustness against mainstream attacks. The protocol reduces area overhead and computational time by 43.04% and 42.99%, respectively, compared to similar protocols.
Physical Unclonable Functions (PUFs) and Hardware Security
This research introduces a post-quantum crypto-infrastructure exploiting the CRYSTALS-Kyber lattice based key encapsulation and the CRYSTALS-Dilithium digital signature cryptography supporting the real time payment transactions against quantum computing based attacks. Besides, based on the open quantum safe library OpenQuantumSafe and TensorFlow Federated, a hybrid classical-quantum federated key management protocol for payment systems is built. With the system, transaction validation can be checked without decryption based on HOMORPHIC encryption using Microsoft SEAL toolkit with sub-200ms latency. For anomaly detection application in neural network, LSTM auto-encoder can detect quantum based attacks with an accuracy of 97.4%. Implementation on AWS Braket Simulator for Verification against Shor algorithm Quantum Resistance and Grover algorithm. The model delivered Redis heavily boronated, for high speed key caching, and Apache Kafka, for asynchronous streaming of transactions at a speed of 50,000 transactions per second. Securely storing the cryptographic keys using an integration with HSM PKCS#11 hardware interface Blockchain anchoring with Hyperledger Fabric: GDPR compliance audit trail immutability with zero-knowledge-proofs (libsnark).
C Ratna Prabha, K Mahesh Babu, Kattinti Dheeraj, N Varsha Priya ¡ 5 authors
"Non-Fungible Tokens" (NFTs) have revolutionized digital ownership by enabling verifiable, unique assets on the blockchain. In this paper, we present an advanced, secure, and scalable "NFT marketplace" built on Ethereum, leveraging Zero-Knowledge Proofs (ZKPs), multi-signature wallets, and AI-driven threat detection to enhance security and decentralization. We implement ERC-721 and ERC-1155 standards, integrate Layer 2 solutions (Polygon zkEVM, Arbitrum) for cost efficiency, and introduce decentralized identity (DID) verification to prevent fraud. Our smart contracts were thoroughly tested using Slither, MythX, and manual audits, while the frontend employs React.js with WalletConnect v2 for cross-wallet compatibility. We also propose a novel royalty enforcement mechanism using ERC2981 with automated payouts via Chainlink Automation. Performance benchmarks demonstrate sub-second transaction finality on Layer 2 and gas cost reductions of up to 90% compared to traditional Ethereum mainnet deployments.
List decoding of codes can be seen as the generalization of unique decoding of codes while list decoding over finite fields has been extensively studied, extending these results to more general algebraic structures such as Galois rings remains an important challenge. Due to recent progress in zero knowledge systems, there is a growing demand to investigate the proximity gap of codes over Galois rings in Yizhou Yao(2025). The proximity gap is closely related to the decoding capability of codes. It was shown in Eli Ben-Sasson(2020) that the proximity gap for RS codes over finite field can be improved to $1-\sqrt{r}$ if one consider list decoding instead of unique decoding. However, we know very little about RS codes over Galois ring which might hinder the development of zero knowledge proof system for ring-based arithmetic circuit. In this work, we first extend the list decoding procedure of Guruswami and Sudan to Reed-Solomon codes over Galois rings, which shows that RS codes with rate $r$ can be list decoded up to radius $1-\sqrt{r}$. Then, we investigate the list decoding of folded Reed-Solomon codes over Galois rings. We show that the list decoding radius of folded Reed-Solomon codes can reach the Singlton bound as its counterpart over finite field. We also extend the deterministic pruning method of Vikrant Ashvinkumar(2026) to Galois rings, showing how to prune the affine free module obtained from the linear-algebraic decoder and recover the candidate codewords. Finally, we improve the list size of our folded Reed-Solomon code to $O(1/\varepsilon^2)$ by extending recent work in Shashank Srivastava(2025) to Galois Rings. By developing the recent work of Yeyuan Chen(2025), we show that folded Reed-Solomon codes over Galois rings satisfy the relaxed generalized Singleton bound in the average-radius sense with optimal list size $O(1/\varepsilon)$.
Blockchain and Federated Learning (FL) provide a strong framework for distributed, privacypreserving machine learning in the medical field. In order to provide safe and effective model training, this framework assists in handling sensitive patient data from lung disease diagnosis, such as CT scans, X-rays, and clinical records. The proposed approach improves distributed machine learning security, privacy, and integrity, particularly in delicate fields like healthcare. Contributions from other datasets help the model get better, but patient data is kept private and blockchain guarantees the integrity of the updates to the model. ZeroKnowledge Proofs (ZKP) guarantee that customers can demonstrate the accuracy of their model upgrades without disclosing any personal information. FLBC- ZKP uses cryptographic proofs to remove this requirement for confidence. FLBC-ZKP models exhibit competitive accuracy rates in healthcare applications, guaranteeing confidentiality and privacy without compromising predictive performance. Contri- butions from other datasets improve the model, but patient information is kept confidential and the blockchain ensures the accuracy of model updates. Compared to regular FL, FLBC-ZKP delivers superior privacy and security through blockchain and ZKP, making it particularly suitable for sensitive healthcare data, while maintaining high accuracy. The accuracy data throughout federated learning rounds for a different approach, FLBC-ZKP slightly surpasses the other methods as the number of rounds increases.
The absence of a fully decentralized, verifiable, and privacy-preserving communication protocol for autonomous agents remains a core challenge in decentralized computing. Existing systems often rely on centralized intermediaries, which reintroduce trust bottlenecks, or lack decentralized identity-resolution mechanisms, limiting persistence and cross-network interoperability. We propose the Decentralized Interstellar Agent Protocol (DIAP), a novel framework for agent identity and communication that enables persistent, verifiable, and trustless interoperability in fully decentralized environments. DIAP binds an agent's identity to an immutable IPFS or IPNS content identifier and uses zero-knowledge proofs (ZKP) to dynamically and statelessly prove ownership, removing the need for record updates. We present a Rust SDK that integrates Noir (for zero-knowledge proofs), DID-Key, IPFS, and a hybrid peer-to-peer stack combining Libp2p GossipSub for discovery and Iroh for high-performance, QUIC based data exchange. DIAP introduces a zero-dependency ZKP deployment model through a universal proof manager and compile-time build script that embeds a precompiled Noir circuit, eliminating the need for external ZKP toolchains. This enables instant, verifiable, and privacy-preserving identity proofs. This work establishes a practical, high-performance foundation for next-generation autonomous agent ecosystems and agent-to-agent (A to A) economies.
Zero-knowledge proof is one of the most promising privacy-preserving approaches in the current literature. However, its complex nature leads its development and deployment to be time-consuming and error-prone. Furthermore, despite the increasing adoption of zero-knowledge proof in the blockchain applications, there is still no systematic framework that streamlines the end-to-end life-cycle of proof development. This paper introduces the first LLM-enhanced zero-knowledge proof DevOps framework for blockchain (i.e. zkOps) to the literature. To evaluate the performance of the framework on different real-life scenarios, a small benchmarking suite is constructed with the increasing computational complexity with respect to the size of circuit constraints. The experimental study identifies the effects of the model temperature on the code compilation rate, and the complexity of prompts on the service latencies. The findings show that the framework efficiently handles the varying-complexity of prompts with a maximum successful compilation rate of 70% (i.e., up to 200,000 proof constraints).
The rapid progress of quantum computing poses significant challenges to traditional cryptographic mechanisms, necessitating the adoption of post-quantum cryptography (PQC) solutions. This paper proposes a Quantum-Enhanced Security for Smart Meters (QESM) system to protect power plant data in smart cities, integrating Kyber for secure key exchange, FALCON (Fast-Fourier Transform over Lattice-based Cryptography) for quantum-resistant digital signatures, and ZKP (Zero-Knowledge Proof) for effective verification without revealing sensitive data to secure power plant data against quantum attacks. To evaluate the security of the proposed system, we analyze its resistance to various quantum threats, including Shorâs algorithm, Groverâs algorithm, quantum key analysis, quantum reversal encryption, quantum amplification, quantum switching, and quantum collision attacks. In the current study, accurate measures were used and the average was approximately 7.065 (bits/byte) for randomness, the average execution time was 6.202 milliseconds, the average memory consumption was approximately 4.343 KB, 6.4 Completeness was equal to 1 and unforgeability was 100%. As for the average throughput, it was approximately 485,605 operations per second. That shows the QESM system provides strong security and efficiency, making it a viable solution for protecting the electricity infrastructure in smart cities in the quantum era.
Barrett's algorithm is one of the most widely used methods for performing modular multiplication, a critical nonlinear operation in modern privacy computing techniques such as homomorphic encryption (HE) and zero-knowledge proofs (ZKP). Since modular multiplication dominates the processing time in these applications, computational complexity and memory limitations significantly impact performance. Computing-in-Memory (CiM) is a promising approach to tackle this problem. However, existing schemes currently suffer from two main problems: 1) Most works focus on low bit-width modular multiplication, which is inadequate for mainstream cryptographic algorithms such as elliptic curve cryptography (ECC) and the RSA algorithm, both of which require high bit-width operations; 2) Recent efforts targeting large number modular multiplication rely on inefficient in-memory logic operations, resulting in high scaling costs for larger bit-widths and increased latency. To address these issues, we propose LaMoS, an efficient SRAM-based CiM design for large-number modular multiplication, offering high scalability and area efficiency. First, we analyze the Barrett's modular multiplication method and map the workload onto SRAM CiM macros for high bit-width cases. Additionally, we develop an efficient CiM architecture and dataflow to optimize large-number modular multiplication. Finally, we refine the mapping scheme for better scalability in high bit-width scenarios using workload grouping. Experimental results show that LaMoS achieves a $7.02\times$ speedup and reduces high bit-width scaling costs compared to existing SRAM-based CiM designs.
The Riemann Hypothesis (RH) has remained one of the most significant unsolved problems in mathematics for over 160 years. This paper posits a novel argument that the resistance of the RH to proof stems not from mathematical intractability, but from a fundamental ontological incompatibility. The hypothesis, we argue, implicitly presupposes a Platonic ontology, wherein infinite sets (such as the set of all non-trivial zeros) exist as complete, static objects accessible to timeless logical inspection. As a counter-framework, we introduce the KnoWellian Universe Theory (KUT), a procedural ontology where mathematical facts do not pre-exist but are continuously rendered into actuality. KUT is founded upon the Axiom of Bounded Infinity (-c > â < c+), which rejects the hierarchy of completed infinities, and operates via a ternary time structure (Past, Instant, Future) that governs the dynamic interplay of Control (actualized reality) and Chaos (unmanifested potential). From these axioms, we derive the Law of KnoWellian Conservation (a(t) + w(t) = N), which formally partitions reality into a finite set of rendered facts, a(t), and a vast, unrendered potential, w(t). We demonstrate that a deductive proof of the RH would require certain knowledge of the properties of the unrendered set w(t), a logical impossibility for any observer existing within the procedural universe. Through the 'Bernharda' thought experiment, we illustrate that any consciousness capable of such a proof would necessarily be a 'Boltzmann Brain'âa mind predicated on the ontologically false Platonic substrate. We conclude that the Riemann Hypothesis is not provably true or false within a KnoWellian framework, but is un-renderable: a beautiful and well-formed question formulated in the language of static 'being' that cannot be answered in a universe of dynamic 'becoming'. The paper includes a formal proof of un-renderability, a discussion of objections and implications, and a comparison between Platonic and KnoWellian (procedural) ontologies, positioning KUT within the historical context of foundational debates in mathematics (e.g., Intuitionism).
Abstract In todayâs era of digital transformation, online transactions have become vital to financial systems, e-commerce, and decentralized applications. However, increasing dependence on digital payment infrastructures has also raised major security concerns such as hacking, identity theft, and unauthorized access. To address these challenges, the proposed project âBlockchain Secure Transactionâ presents a decentralized framework that ensures transparency, integrity, and confidentiality in digital transactions. The system uses blockchain technology to record and validate each transaction in a distributed ledger, eliminating centralized control and making data immutable and tamper-proof. The workflow begins with user registration, where users provide details and set a picture password for secure recognition. During login, the system verifies credentials and performs biometric authentication to confirm user identity. Unregistered users are redirected to the registration page, maintaining process integrity. Once authenticated, users access the dashboard to initiate secure transactions. To preserve privacy, Zero-Knowledge Proof (ZKP) is used, allowing users to prove transaction authenticity without revealing sensitive information. Transactions then pass through smart contract verification, which ensures compliance with predefined conditions. Successful verifications result in completed transactions, while suspicious or invalid ones are blocked or frozen automatically. All user data and transaction logs are securely stored in Firebase, with backend processing handled in Java and the frontend designed using React (app.jsx). By combining blockchainâs immutability, smart contract automation, ZKP privacy proofs, and biometric authentication, the Blockchain Secure Transaction System offers a multi-layered, tamper-resistant, and transparent solution for secure online payments â enhancing trust and reliability in the digital economy.
Privacy in blockchain-based cryptocurrencies has become a critical area of research and development. This is driven by the transparent nature of public ledgers and the growing demand for confidential transactions. This study provides a comprehensive exposition of a prominent privacyenhancing protocol, Zcash. We begin by elucidating the design and operation of Tornado Cash, a smart contract-based mixer on Ethereum that leverages zero-knowledge proofs to enable unlinkable transactions. Building on this foundation, we explore the Zcash protocol, tracking its evolution from Bitcoin and its innovative use of zk-SNARKs. We also examine the intricate mechanisms that highlight its privacy guarantees, including circuit design, Merkle tree structures and transaction relations. By systematically analyzing and comparing these protocols, this study aims to clarify their inner workings, highlight their strengths and limitations, and provide information on the future of privacy in decentralized finance.
Pedro Baptista, Bernardo Pacheco, Filipe Apolinårio, João Silveira ¡ 6 authors
Having the ability to prove your knowledge is essential for obtaining a job. In the programming field, applicants make claims about the programming languages they master, and it is up to the interviewer to check the veracity of those claims. The goal of this work is to facilitate this process by extracting important information from GitHub, such as the number of bytes programmed in each programming language. The user is then able to ask for a Zero Knowledge Proof which can be downloaded and sent to any entity which places trust on out platform. The proof is verifiable without the entity interacting with our platform and does not leak information about the users' GitHub. The obtained results are promising, even though proofs can take several minutes to generate, they can be verified in many devices, such as laptops and smartphones, which greatly increases the number of users who can use our platform.
AI-based Problem Solving and Planning
Logic, Reasoning, and Knowledge
Intelligent Tutoring Systems and Adaptive Learning
This research introduces a next-generation cryptographic framework aimed at securing payment systems and IoT-enabled financial ecosystems against both contemporary cyber threats and the anticipated risks posed by quantum computing. The proposed architecture supports intelligent payment cards, contactless transactions, and IoT-based banking infrastructures by integrating classical cryptographic methodsâsuch as Advanced Encryption Standard (AES), RivestâShamirâAdleman (RSA), Elliptic Curve Cryptography (ECC), Secure Hash Algorithm 3 (SHA-3), and Hash-based Message Authentication Code (HMAC)âwith post-quantum algorithms, including CRYSTALS-Kyber and Dilithium. In addition, the framework aligns with widely adopted industry standards, such as the Payment Card Industry Data Security Standard (PCI DSS), Europay-MasterCard-Visa (EMV), ISO 27001, and National Institute of Standards and Technology (NIST) guidelines, thereby ensuring compliance and regulatory resilience. To address the evolving cyber threat landscape, the system integrates blockchain-based decentralized identity management with zero-knowledge proofs (ZKPs) for trustless authentication. Furthermore, privacy-preserving techniques such as Secure Multi-Party Computation (MPC) and Fully Homomorphic Encryption (FHE) are employed to enable secure processing of encrypted data. The security framework is further strengthened through AI-driven fraud detection, which leverages deep learning and federated learning models to detect anomalies in real time without compromising customer privacy. Additionally, confidential computing enclaves and hardware trust anchorsâincluding Hardware Security Modules (HSMs), Trusted Platform Modules (TPMs), and Physically Unclonable Functions (PUFs)âare utilized to enhance system integrity and reliability. By combining quantum-resistant cryptography, privacy-preserving computation, and intelligent anomaly detection, this work presents an adaptive and future-proof security roadmap for the financial sector. The proposed framework is designed not only to counteract todayâs sophisticated cyberattacks but also to maintain resilience against the emerging challenges of the quantum era.
Physical Unclonable Functions (PUFs) and Hardware Security
The emergence of 6G-connected smart cities introduces unprecedented challenges in ensuring security, privacy, and scalability for billions of heterogeneous devices and mission-critical services. Traditional Zero Trust Architectures (ZTA) provide continuous verification but rely on centralized control, making them vulnerable to insider threats and single points of failure. Conversely, blockchain-based frameworks ensure immutability and decentralized trust but suffer from high latency and limited scalability. This paper proposes a novel Blockchain-Enabled Zero Trust Architecture (BZTA) that integrates blockchainâs distributed trust management with Zero Trustâs continuous authentication and micro-segmentation, optimized for 6G urban infrastructures.The contributions of this work are fourfold. First, we design a four-layer BZTA model incorporating decentralized identity management, Zero Trust Gateways (ZTGs), blockchain-based ledgers, and smart contracts for adaptive access control. Second, we formalize the methodological foundations of the framework, including trust computation, Zero-Knowledge Proof (ZKP)-based authentication, authorization logic, and Proof-of-Authority consensus mechanisms. Third, we present a comprehensive evaluation using analytical models and simulations. Results show that BZTA achieves 95% trust classification accuracy, sub-20 ms authentication latency compliant with 6G URLLC, revocation within 3 s, and throughput up to 50,000 transactions per second, while reducing authentication energy costs by 35â40% compared to blockchain-only systems. Fourth, we demonstrate that BZTA provides robust defense against spoofing, replay, insider, lateral, and location spoofing attacks, significantly outperforming both ZTA-only and blockchain-only approaches.The findings highlight BZTA as a scalable, resilient, and privacy-preserving security paradigm for 6G smart cities. By merging blockchain immutability with Zero Trustâs dynamic verification, BZTA enables secure and transparent deployment of critical urban services such as healthcare, transportation, and energy management. This work positions BZTA as a foundational step toward building resilient, citizen-centric, and quantum-ready smart city infrastructures.
Nurhajar Anugraha, Muhammad Riswanto, Lindawati Lindawati, Asrul Asrul
Penelitian ini bertujuan untuk mengembangkan sistem e-voting berbasis blockchain dengan autentikasi biometrik sidik jari serta penerapan protokol zero-knowledge proofs sebagai pengamanan tambahan terhadap data pemilih dan hasil suara. Permasalahan utama yang dihadapi dalam sistem pemungutan suara elektronik konvensional adalah rendahnya kepercayaan terhadap keamanan data dan potensi manipulasi hasil. Metode penelitian yang digunakan mencakup perancangan sistem dengan arsitektur client-server, implementasi teknologi blockchain untuk pencatatan suara yang terenkripsi, serta integrasi biometrik sidik jari menggunakan BiometricPrompt API pada Android. Selain itu, sistem diverifikasi dengan kode OTP melalui email institusional sebagai bentuk validasi ganda pengguna. Hasil pengujian menunjukkan bahwa sistem dapat berjalan dengan baik dan memberikan keamanan yang tinggi karena setiap data suara tersimpan secara permanen dan tidak dapat diubah di jaringan blockchain. Autentikasi biometrik juga memastikan bahwa setiap pemilih terverifikasi secara unik sehingga tidak terjadi pemungutan suara ganda. Dengan demikian, sistem e-voting ini dinilai layak diterapkan untuk lingkungan akademik dan dapat dikembangkan lebih lanjut untuk pemilihan umum berskala lebih besar.
As the "agentic web" takes shape-billions of AI agents (often LLM-powered) autonomously transacting and collaborating-trust shifts from human oversight to protocol design. In 2025, several inter-agent protocols crystallized this shift, including Google's Agent-to-Agent (A2A), Agent Payments Protocol (AP2), and Ethereum's ERC-8004 "Trustless Agents," yet their underlying trust assumptions remain under-examined. This paper presents a comparative study of trust models in inter-agent protocol design: Brief (self- or third-party verifiable claims), Claim (self-proclaimed capabilities and identity, e.g. AgentCard), Proof (cryptographic verification, including zero-knowledge proofs and trusted execution environment attestations), Stake (bonded collateral with slashing and insurance), Reputation (crowd feedback and graph-based trust signals), and Constraint (sandboxing and capability bounding). For each, we analyze assumptions, attack surfaces, and design trade-offs, with particular emphasis on LLM-specific fragilities-prompt injection, sycophancy/nudge-susceptibility, hallucination, deception, and misalignment-that render purely reputational or claim-only approaches brittle. Our findings indicate no single mechanism suffices. We argue for trustless-by-default architectures anchored in Proof and Stake to gate high-impact actions, augmented by Brief for identity and discovery and Reputation overlays for flexibility and social signals. We comparatively evaluate A2A, AP2, ERC-8004 and related historical variations in academic research under metrics spanning security, privacy, latency/cost, and social robustness (Sybil/collusion/whitewashing resistance). We conclude with hybrid trust model recommendations that mitigate reputation gaming and misinformed LLM behavior, and we distill actionable design guidelines for safer, interoperable, and scalable agent economies.
Vivi Andersson, Sofia Bobadilla, Harald Hobbelhagen, Martin Monperrus
Smart contracts operate in a highly adversarial environment, where vulnerabilities can lead to substantial financial losses. Thus, smart contracts are subject to security audits. In auditing, proof-of-concept (PoC) exploits play a critical role by demonstrating to the stakeholders that the reported vulnerabilities are genuine, reproducible, and actionable. However, manually creating PoCs is time-consuming, error-prone, and often constrained by tight audit schedules. We introduce PoCo, an agentic framework that automatically generates executable PoC exploits from natural-language vulnerability descriptions written by auditors. PoCo autonomously generates PoC exploits in an agentic manner by interacting with a set of code-execution tools in a ReasonâActâObserve loop. It produces fully executable exploits compatible with the Foundry testing framework, ready for integration into audit reports and other security tools. We evaluate PoCo on a dataset of 23 real-world vulnerability reports. PoCo consistently outperforms the Zero-shot and Workflow baselines, generating well-formed and logically correct PoCs. Our results demonstrate that agentic frameworks can significantly reduce the effort required for high-quality PoCs in smart contract audits. Our contribution provides actionable knowledge for the smart contract security community.