SPEAR: An Engineering Case Study of Multi-Agent Coordination for Smart Contract Auditing

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chebolu, Indraveni, Mallick, Arnab, Rana, Harmesh
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914462432231424
author Chebolu, Indraveni
Mallick, Arnab
Rana, Harmesh
author_facet Chebolu, Indraveni
Mallick, Arnab
Rana, Harmesh
contents We present SPEAR, a multi-agent coordination framework for smart contract auditing that applies established MAS patterns in a realistic security analysis workflow. SPEAR models auditing as a coordinated mission carried out by specialized agents: a Planning Agent prioritizes contracts using risk-aware heuristics, an Execution Agent allocates tasks via the Contract Net protocol, and a Repair Agent autonomously recovers from brittle generated artifacts using a programmatic-first repair policy. Agents maintain local beliefs updated through AGM-compliant revision, coordinate via negotiation and auction protocols, and revise plans as new information becomes available. An empirical study compares the multi-agent design with centralized and pipeline-based alternatives under controlled failure scenarios, focusing on coordination, recovery behavior, and resource use.
format Preprint
id arxiv_https___arxiv_org_abs_2602_04418
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SPEAR: An Engineering Case Study of Multi-Agent Coordination for Smart Contract Auditing
Chebolu, Indraveni
Mallick, Arnab
Rana, Harmesh
Multiagent Systems
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
Emerging Technologies
Software Engineering
We present SPEAR, a multi-agent coordination framework for smart contract auditing that applies established MAS patterns in a realistic security analysis workflow. SPEAR models auditing as a coordinated mission carried out by specialized agents: a Planning Agent prioritizes contracts using risk-aware heuristics, an Execution Agent allocates tasks via the Contract Net protocol, and a Repair Agent autonomously recovers from brittle generated artifacts using a programmatic-first repair policy. Agents maintain local beliefs updated through AGM-compliant revision, coordinate via negotiation and auction protocols, and revise plans as new information becomes available. An empirical study compares the multi-agent design with centralized and pipeline-based alternatives under controlled failure scenarios, focusing on coordination, recovery behavior, and resource use.
title SPEAR: An Engineering Case Study of Multi-Agent Coordination for Smart Contract Auditing
topic Multiagent Systems
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
Emerging Technologies
Software Engineering
url https://arxiv.org/abs/2602.04418