Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Arnab Mallick, Indraveni Chebolu, Harmesh Rana

TL;DR

The paper tackles the scalability and reliability challenge of security auditing for smart contracts by introducing SPEAR, a multi-agent framework that coordinates specialized agents ($A_P$, $A_E$, $A_R$, $A_C$, $A_{Coord}$) using Contract Net, plan negotiation, and resource auctions. Agents maintain AGM-compliant beliefs, autonomously revise plans, and collaborate to allocate tasks and repair brittle artifacts, enabling robust, resource-aware auditing in dynamic, failure-prone environments. The authors provide an engineering case study with a risk-aware planning agent, programmatic-first PFIR self-healing, and a thorough empirical evaluation against centralized and pipeline baselines, showing improved effectiveness, robustness, and efficiency. This work demonstrates that established MAS coordination patterns can be composed to manage long-running, heterogeneous tooling pipelines for smart contract analysis, with practical implications for faster vulnerability discovery and reduced reliance on costly human intervention.

Abstract

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.

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

TL;DR

The paper tackles the scalability and reliability challenge of security auditing for smart contracts by introducing SPEAR, a multi-agent framework that coordinates specialized agents (, , , , ) using Contract Net, plan negotiation, and resource auctions. Agents maintain AGM-compliant beliefs, autonomously revise plans, and collaborate to allocate tasks and repair brittle artifacts, enabling robust, resource-aware auditing in dynamic, failure-prone environments. The authors provide an engineering case study with a risk-aware planning agent, programmatic-first PFIR self-healing, and a thorough empirical evaluation against centralized and pipeline baselines, showing improved effectiveness, robustness, and efficiency. This work demonstrates that established MAS coordination patterns can be composed to manage long-running, heterogeneous tooling pipelines for smart contract analysis, with practical implications for faster vulnerability discovery and reduced reliance on costly human intervention.

Abstract

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.
Paper Structure (34 sections, 3 figures, 3 tables)

This paper contains 34 sections, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Smart Contract Analysis Tools
  4. MAS Coordination Mechanisms
  5. Self-Healing and Self-Adaptive Systems
  6. Multi-Agent Planning
  7. The SPEAR Framework
  8. Formal Model and Coordination
  9. System State $\mathcal{S}$
  10. Events $\mathcal{E}$
  11. Agent Model
  12. Actions and Execution
  13. Agent Execution Semantics
  14. Coordination Protocols
  15. System Properties
  16. ...and 19 more sections

Figures (3)

  • Figure 1: SPEAR architecture: five specialized agents ($A_P$, $A_E$, $A_R$, $A_C$, $A_{Coord}$) interact through a message bus, coordinating via Contract Net, plan negotiation, and resource auctions.
  • Figure 2: Performance of the PFIR self-healing algorithm. The programmatic-first echelon resolves most failures, achieving consistent success across batches.
  • Figure 3: Impact of the Strategic Planning Agent. The planning-driven mode detects critical vulnerabilities faster than the ablated baseline.