Glossary
Short, sourced definitions of the terms used in security operations, detection, and AI security.
A
- AI agentAgentic AI · LLM agentAn AI agent is a system in which a large language model dynamically directs its own process and tool use to complete a task, deciding which steps to take rather than following a fixed, predefined script.
- AI hallucinationLLM hallucination · ConfabulationAn AI hallucination is output from a generative AI model that is false, fabricated, or inconsistent with its inputs, but presented confidently as if it were accurate.
- AI red teamingLLM red teaming · GenAI red teamingAI red teaming is the practice of adversarially testing AI models and AI-enabled systems to find harmful or unintended behavior, security vulnerabilities such as prompt injection, and weaknesses in safeguards before attackers or users do.
- Alert triageSecurity alert triage · SOC triageAlert triage is the first review of a security alert, in which an analyst or automated system confirms what happened, judges whether it is malicious and how severe it is, and decides to close it, escalate it, or start a response.
B
- Blue teamBlue teaming · Defensive security teamA blue team is the group responsible for defending an organization's systems by maintaining its security posture and detecting and responding to attacks, whether real or simulated by a red team.
- Business email compromise (BEC)BEC · Email account compromiseBusiness email compromise (BEC) is a scam in which attackers compromise or impersonate a legitimate business email account and use it to trick employees, customers, or partners into transferring funds or handing over sensitive information.
C
- CIEMCloud infrastructure entitlement management · Cloud entitlement managementCIEM (cloud infrastructure entitlement management) is the practice and tooling for analyzing and managing the identities, permissions, and entitlements in cloud environments, to find excessive access and enforce least privilege.
- CNAPPCloud-native application protection platform · Cloud native application protection platformA CNAPP (cloud-native application protection platform) is a security platform that combines cloud posture management, identity and entitlement management, and workload protection to secure cloud-native applications from development through runtime.
- Command and control (C2)C2 · C&CCommand and control (C2) is the communication channel attackers use to send instructions to systems they have compromised and receive results, usually disguised to look like normal network traffic.
- Credential stuffingCredential stuffing attack · T1110.004Credential stuffing is an automated attack in which username and password pairs stolen in one breach are tried against login forms on other services, exploiting the fact that many people reuse passwords.
- CSPMCloud security posture management · Cloud posture managementCSPM (cloud security posture management) is the practice and tooling for continuously assessing cloud infrastructure configurations against security standards and best practices, to find and remediate misconfigurations and compliance violations.
- Cyber kill chainIntrusion kill chain · Lockheed Martin Cyber Kill ChainThe cyber kill chain is a model, published by Lockheed Martin researchers, that breaks an intrusion into seven sequential phases so defenders can detect and interrupt an attacker at any step before they reach their objective.
D
- Data exfiltrationExfiltration · Data theftData exfiltration is the unauthorized transfer of data out of an organization's environment, whether by an external attacker or an insider, typically disguised as normal traffic or sent to services the organization already uses.
- Data loss prevention (DLP)DLP · Data leak preventionData loss prevention (DLP) is the set of policies and tools that identify sensitive data and detect or block its unauthorized use, sharing, or transfer, whether the data is in use on endpoints, in motion over networks, or at rest in storage.
- Detection engineeringDetection development · Detection engineerDetection engineering is the practice of designing, building, testing, deploying, and maintaining the rules and analytics that turn security telemetry into alerts, treated with the rigor of software engineering.
- DFIRDigital forensics and incident response · Digital forensicsDFIR (digital forensics and incident response) is the discipline of investigating security incidents by collecting and analyzing digital evidence to establish what happened, while containing the attack and restoring affected systems.
E
- EDREndpoint detection and response · Endpoint detection and threat responseEDR (endpoint detection and response) is security software that records activity on laptops, servers, and other endpoints, detects suspicious behavior in that telemetry, and gives responders tools to investigate and contain affected hosts.
- Elastic Common SchemaECSThe Elastic Common Schema (ECS) is an open-source specification that defines common field names, data types, and descriptions for storing logs, metrics, and security events, so data from different sources can be analyzed and correlated together.
F
I
- Incident responseIR · Cyber incident responseIncident response is the set of processes an organization uses to detect, analyze, contain, eradicate, and recover from cybersecurity incidents, and to feed what it learns back into its defenses.
- Incident response playbookSecurity playbook · IR playbookAn incident response playbook is a documented set of actionable steps for handling a specific type of security incident or alert, such as phishing, ransomware, or a compromised account, so responders act consistently and quickly.
- Indicators of compromiseIOC · IoCIndicators of compromise (IOCs) are technical artifacts, such as file hashes, IP addresses, domain names, and URLs, that suggest an attack is underway or that a system has already been compromised.
- ITDRIdentity threat detection and response · Identity threat detectionITDR (identity threat detection and response) is the practice and tooling for detecting, investigating, and stopping attacks that abuse identities, such as compromised credentials, privilege escalation, and lateral movement, using telemetry from identity providers and directories.
K
L
- Lateral movementLateral spread · PivotingLateral movement is the set of techniques attackers use to move from a system they have compromised to other systems and accounts on the same network, pivoting toward the data or access they ultimately want.
- Living off the land (LOTL)LOTL · LOLBinsLiving off the land (LOTL) is an attack approach in which adversaries use legitimate tools and features already present on a system, such as PowerShell, WMI, or standard Unix utilities, instead of their own malware, so their activity blends in with normal administration.
- Log normalizationLog parsing · Data normalizationLog normalization is the process of converting each field in log data from many different sources into a consistent name, format, and data type, so events can be searched, correlated, and analyzed together.
M
- MDRManaged detection and responseMDR (managed detection and response) is a security service in which an outside provider's analysts monitor an organization's environment around the clock, investigate alerts, hunt for threats, and take or guide containment actions.
- MITRE ATT&CKATT&CK · MITRE ATT&CK frameworkMITRE ATT&CK is a free, publicly maintained knowledge base that catalogs the tactics and techniques attackers use against real systems, based on observed intrusions, and gives defenders a shared vocabulary for detection and threat intelligence.
- Model Context Protocol (MCP)MCP · MCP serverThe Model Context Protocol (MCP) is an open standard that defines how AI applications connect to external tools, data sources, and workflows, so a model can read context and take actions through a common interface.
P
- Privilege escalationPrivesc · Elevation of privilegePrivilege escalation is the set of techniques attackers use to gain higher-level permissions on a system or network than they initially had, such as moving from a standard user to administrator, root, or a privileged cloud role.
- Prompt injectionIndirect prompt injection · Prompt injection attackPrompt injection is an attack in which text supplied to a large language model, directly by a user or hidden in content the model reads, alters its behavior in ways its developers didn't intend, such as ignoring instructions, leaking data, or misusing tools.
- Purple teamPurple teaming · Purple team exercisePurple teaming is a collaborative security exercise in which red team attackers and blue team defenders work together, emulating specific adversary behaviors and testing and improving detection and response as they go.
R
S
- Security operations centerSOC · Security operationsA security operations center (SOC) is the team, processes, and technology an organization uses to continuously monitor its systems and to detect, investigate, and respond to cybersecurity threats.
- Shadow AIUnsanctioned AI · Unauthorized AI useShadow AI is the use of AI tools, models, or services by employees without the approval or oversight of the organization's IT or security teams.
- SIEMSecurity information and event managementA SIEM (security information and event management) system collects security logs from across an organization, stores them in one searchable place, and runs detection rules over them so a security team can find, investigate, and respond to threats.
- Sigma rulesSigma · Sigma ruleSigma is an open, vendor-neutral format for writing detection rules for log data in YAML, so a rule can be written once, shared, and converted into the query language of whichever SIEM or log platform a team uses.
- SOC analystSecurity operations center analyst · Security analystA SOC analyst is a security professional who monitors alerts from an organization's security tools, investigates suspicious activity, and decides whether to close, escalate, or contain each potential threat.
- SPLSearch Processing Language · Splunk Search Processing LanguageSPL (Search Processing Language) is Splunk's query language, made of search commands chained with pipes that retrieve events from indexes and then filter, transform, and aggregate them.
- Supply chain attackSupply chain compromise · Software supply chain attackA supply chain attack compromises a product, service, or supplier that organizations trust, such as a software dependency, build system, update channel, or IT provider, in order to reach the supplier's customers.
T
- Threat intelligenceCyber threat intelligence · CTIThreat intelligence is information about threats and threat actors that has been collected, analyzed, and enriched with context so defenders can make decisions, such as what to detect, hunt for, patch, or block.
- Threat modelingThreat model · STRIDEThreat modeling is the practice of analyzing a representation of a system to identify what could go wrong with its security and privacy, and deciding which countermeasures to apply.
- TTPsTactics, techniques, and proceduresTTPs (tactics, techniques, and procedures) describe how an attacker behaves, from the high-level goal of an action (tactic) to the method used to reach it (technique) to the specific way a particular actor carries it out (procedure).
U
X
Y
Z
Looking for more than a definition? Learn has full guides, pricing breakdowns, and comparisons.
Attackers are scaling with tokens
Cotool helps defenders operate at machine speed. See how security teams are scaling Detection & Response.
Book a demo