Skip to main content

Glossary

Comprehensive definitions of key terms and concepts used throughout BevorAI documentation and the broader blockchain security ecosystem.

BevorAI-Specific Terms

BevorAI

The AI-powered security auditing platform that provides comprehensive security analysis, real-time monitoring, and automated threat detection for blockchain applications and smart contracts.

Security Agent

Specialized AI agents within the BevorAI ecosystem that focus on specific security domains (e.g., reentrancy detection, economic analysis, governance security). Agents can work independently or in coordination with other agents.

Multi-Agent Triaging (A2A)

Agent-to-Agent coordination system where multiple specialized security agents collaborate to analyze complex security scenarios, prioritize findings, and provide comprehensive risk assessments.

MCP Integration

Model Context Protocol integration that enables BevorAI to work directly within development environments like Cursor and Windsurf, providing real-time security analysis as developers write code.

Security Data Labeling

The process of expert security auditors reviewing and tagging BevorAI’s findings to improve the accuracy and effectiveness of the AI models. Contributors earn rewards for high-quality labeling.

BEVOR Token

The native utility token of the BevorAI ecosystem, used for governance, accessing premium features, rewarding contributors, and incentivizing security improvements across the network.

Blockchain Security Terms

Security Audit

A comprehensive review of smart contract code, protocol design, and system architecture to identify vulnerabilities, assess risks, and provide remediation recommendations.

Auditor

A security professional who specializes in reviewing blockchain code and protocols to identify vulnerabilities and security risks. Auditors may work independently or as part of auditing firms.

Smart Contract

Self-executing contracts with terms directly written into code. Smart contracts run on blockchain networks and automatically enforce agreements without intermediaries.

Penetration Testing

Simulated cyber attacks against smart contracts or protocols to identify exploitable vulnerabilities and assess the security posture of the system.

Vulnerability Types

Reentrancy Attack

A vulnerability where an external contract calls back into the calling contract before the first function call is finished, potentially allowing attackers to drain funds or manipulate state.
// Vulnerable to reentrancy
function withdraw() external {
    uint amount = balances[msg.sender];
    msg.sender.call{value: amount}(""); // External call
    balances[msg.sender] = 0; // State change after external call
}

Access Control Vulnerability

Security issues related to improper permission management, where unauthorized users can access restricted functions or escalate privileges.

Oracle Manipulation

Attacks that exploit price oracles or data feeds to manipulate protocol behavior, often through flash loans or coordinated market manipulation.

Flash Loan Attack

Exploits that use uncollateralized loans (flash loans) to manipulate protocol state, extract value, or exploit economic vulnerabilities within a single transaction.

Front-Running

The practice of observing pending transactions and submitting competing transactions with higher gas fees to execute first, potentially extracting MEV (Maximal Extractable Value).

Governance Attack

Exploits targeting governance mechanisms, including vote buying, flash governance (using borrowed tokens to vote), and proposal manipulation.

Technical Concepts

Gas Optimization

The process of reducing the computational cost (gas) required to execute smart contract functions, improving efficiency and reducing user costs.

Formal Verification

Mathematical techniques used to prove that smart contracts behave correctly according to their specifications, providing higher assurance than testing alone.

Economic Security

Security measures that rely on economic incentives and disincentives to prevent attacks, such as staking mechanisms and slashing conditions.

Slippage

The difference between expected and actual transaction prices, often exploited in DeFi protocols through sandwich attacks and price manipulation.

MEV (Maximal Extractable Value)

The maximum value that can be extracted from block production beyond standard block rewards, often through transaction reordering, inclusion, or exclusion.

TVL (Total Value Locked)

The total amount of cryptocurrency locked in a DeFi protocol, often used as a measure of protocol adoption and security importance.

DeFi-Specific Terms

Automated Market Maker (AMM)

A type of decentralized exchange that uses algorithmic trading to provide liquidity and determine asset prices based on mathematical formulas.

Liquidity Pool

Collections of tokens locked in smart contracts that provide liquidity for decentralized trading and earn fees for liquidity providers.

Yield Farming

The practice of lending or staking cryptocurrency to earn rewards, often involving complex strategies across multiple DeFi protocols.

Impermanent Loss

The temporary loss of funds experienced by liquidity providers due to volatility in trading pairs compared to simply holding the assets.

Liquidation

The process of forcefully selling collateral when a borrower’s position falls below required collateral ratios, ensuring protocol solvency.

Governance Terms

DAO (Decentralized Autonomous Organization)

An organization governed by smart contracts and token holders rather than traditional management structures, using blockchain-based voting mechanisms.

Governance Token

Cryptocurrency tokens that give holders voting rights in protocol governance decisions, such as parameter changes, upgrades, and treasury management.

Proposal

A formal suggestion for changes to a protocol, submitted to the governance system for community voting and potential implementation.

Quorum

The minimum number of votes required for a governance proposal to be valid and executable.

Timelock

A security mechanism that delays the execution of approved governance proposals, providing time for the community to respond to malicious proposals.

Network and Infrastructure

Layer 2 (L2)

Blockchain scaling solutions built on top of main blockchain networks (Layer 1) to increase transaction throughput and reduce costs.

Rollup

A type of Layer 2 scaling solution that processes transactions off-chain and submits compressed data to the main chain for security and finality.

Bridge

Smart contracts that enable the transfer of assets and data between different blockchain networks, often targets for security attacks due to their complexity.

Validator

Entities responsible for validating transactions and maintaining blockchain network security, often through proof-of-stake mechanisms.

Consensus Mechanism

The protocol used by blockchain networks to agree on the validity of transactions and the state of the blockchain.

Analysis and Monitoring

Real-Time Analysis

Continuous security analysis that processes code changes, transactions, or system events as they occur, providing immediate feedback and alerts.

Static Analysis

Code analysis performed without executing the program, examining code structure, patterns, and potential vulnerabilities through automated tools.

Dynamic Analysis

Security analysis performed during program execution, observing actual behavior, state changes, and runtime characteristics.

Threat Intelligence

Information about current and emerging security threats, attack patterns, and vulnerabilities relevant to blockchain and DeFi ecosystems.

Incident Response

Systematic approach to handling security incidents, including detection, containment, analysis, recovery, and lessons learned.

Risk Assessment

Risk Score

Numerical assessment of security risk, typically ranging from 0-100, based on vulnerability severity, exploitability, and potential impact.

False Positive

A security alert or finding that incorrectly identifies benign code as vulnerable, requiring human review to filter out incorrect results.

False Negative

A failure to detect an actual vulnerability, representing a gap in security analysis that could leave systems exposed to attacks.

Attack Vector

A specific method or pathway that an attacker could use to exploit a vulnerability and compromise a system.

Threat Model

A structured approach to identifying and analyzing potential threats, including threat actors, attack methods, and system vulnerabilities.

Integration Terms

API (Application Programming Interface)

A set of protocols and tools for building software applications, allowing different systems to communicate and share data.

Webhook

HTTP callbacks that allow applications to receive real-time notifications when specific events occur, enabling automated responses and integrations.

SDK (Software Development Kit)

Deprecated in these docs. Use our API and webhook integrations.

CI/CD (Continuous Integration/Continuous Deployment)

Development practices that involve automatically building, testing, and deploying code changes to ensure rapid, reliable software delivery.

Rate Limiting

Controls that limit the number of API requests a user can make within a specific time period to prevent abuse and ensure fair resource allocation.

Need More Definitions?

Don’t see a term you’re looking for? We’re constantly expanding our glossary based on community needs.
This glossary is maintained by the BevorAI community. Definitions are regularly updated to reflect current best practices and emerging concepts in blockchain security.
I