CyberKit

Glossary

A glossary of cybersecurity and DFIR terms.

The vocabulary the rest of the site is written in. Each term has its own page with a definition, why it matters, an example, and the tools that work with it.

The core terms and the AI and automation security set go further: what to look for in evidence, the mistake people most often make, and a short list of primary sources from NIST, CISA, NSA and MITRE.

569 terms in 15 categories.

Memory Forensics

Host & Disk Forensics

  • Master File Table The NTFS index of every file, and a goldmine of timestamps.
  • Prefetch A Windows artifact proving a program ran, and when.
  • Shellbags Registry traces of which folders a user browsed in Explorer.
  • Alternate Data Streams Hidden NTFS streams that stash data behind a normal-looking file.
  • Timestomping Forging file timestamps to hide when something really happened.
  • Windows Event Log Windows' record of logons, process starts, and system activity.
  • File Carving Recovering files from raw bytes using signatures, not the file system.
  • USN Journal NTFS log of per-file change records, used to reconstruct activity.
  • $LogFile (NTFS) NTFS journal of pending metadata transactions used to keep the volume consistent.
  • INDX Attribute ($I30) B-tree index NTFS uses for directory listings; its slack can resurrect deleted names.
  • Resident vs. Non-Resident Attributes Whether a file's data lives inside its MFT record or out in disk clusters.
  • MFT Record Slack Unused tail bytes inside a 1 KB MFT record that can retain old metadata.
  • $Bitmap (NTFS) NTFS map of which clusters are allocated; defines what counts as unallocated.
  • Data Runs Compact NTFS map of which clusters hold a non-resident file's content.
  • File Slack Leftover space between a file's real end and its last allocated cluster.
  • Unallocated Space Disk areas not currently assigned to a file but often holding recoverable data.
  • Deleted-File Recovery Restoring removed files from lingering metadata or residual cluster data.
  • Bit-for-Bit Imaging An exact sector-level duplicate of storage media for examination.
  • Write Blocking Hardware or software that prevents any writes to evidence media during acquisition.
  • Image Hash Verification Hashing source and image to prove the forensic copy is bit-identical.
  • Forensic Image Formats Container formats for disk images, from flat raw to compressed, hashed evidence files.
  • Volume Shadow Copy Windows point-in-time volume snapshots that preserve earlier versions of files.
  • Registry Hives On-disk files backing the Windows registry, each holding distinct forensic value.
  • Registry Run Keys Registry locations that auto-launch programs at logon or boot.
  • AmCache Registry hive recording executed/installed programs with hashes and timestamps.
  • ShimCache SYSTEM-hive cache of executable paths and file times, evidence of presence.
  • SRUM Windows database logging per-app resource and network usage over time.
  • Jump Lists Per-application taskbar history of recently and frequently opened items.
  • LNK Files Windows shortcut files that embed rich metadata about their target.
  • Recycle Bin ($I / $R) Per-user deleted-file store pairing metadata ($I) with recoverable content ($R).
  • Thumbnail Cache Cached image/document thumbnails that persist after the originals are deleted.
  • Browser Artifacts On-disk web history, cache, cookies, and downloads from a browser profile.
  • Windows Search Index Indexed properties and text content of files, often outliving the originals.
  • Pagefile On-disk virtual-memory backing store that can hold fragments of RAM.
  • MRU Lists Registry records of recently opened, saved, or typed items per user.
  • UserAssist Per-user registry record of GUI program launches, counts, and last-run time.
  • BAM / DAM Registry record of program execution per user with last-run timestamps.
  • Scheduled-Task Artifacts On-disk task definitions and registry entries used for execution and persistence.
  • Partition Scheme (MBR / GPT) Disk layout metadata defining where volumes begin, end, and what type they are.
  • File Signatures Header/footer byte patterns that identify a file's true type regardless of name.
  • MAC(b) Times The Modified, Accessed, metadata-Changed, and Born timestamps of a file.
  • Timeline Analysis Ordering filesystem and artifact timestamps to reconstruct events.
  • Super Timeline A unified timeline merging filesystem, registry, log, and artifact events.
  • Slack-Space Hiding Concealing data in slack, unused, or non-standard disk areas to evade detection.
  • Live Response Collecting volatile evidence from a running system before it's shut down or imaged.

Malware Analysis & RE

  • Process Hollowing Malware starts a legit process, then swaps its guts for malicious code.
  • Entropy A randomness score that flags packed or encrypted data.
  • Packing Compressing or encrypting a binary to hide its real code.
  • Sandboxing Detonating a sample in an isolated environment to watch its behavior.
  • Rootkit Stealth malware that hides itself deep in the OS.
  • Static Analysis Examining a sample's code and structure without executing it.
  • Dynamic Analysis Running a sample under observation to record what it actually does.
  • Disassembly Translating machine code back into human-readable assembly.
  • Debugging Controlling a running sample step by step to watch its state.
  • Portable Executable Format The Windows file format for executables, DLLs, and drivers.
  • PE Section A named region of a PE holding code, data, or resources.
  • Import Address Table The table the loader fills with addresses of imported functions.
  • String Analysis Extracting readable text from a binary to infer its purpose.
  • Obfuscation Deliberately making code hard to read or analyze.
  • Crypter A tool that encrypts a payload and decrypts it in memory at runtime.
  • Manual Unpacking Reversing a packer by hand to recover the real payload.
  • Polymorphism Malware that changes its appearance each time while keeping behavior.
  • Control-Flow Flattening Obfuscation that hides logic behind a dispatcher and state variable.
  • String Stacking Building strings byte by byte at runtime to hide them from triage.
  • Dynamic API Resolution Looking up needed functions at runtime instead of importing them.
  • API Hashing Referring to functions by hash to hide which APIs malware uses.
  • Direct System Calls Invoking the kernel directly to skip monitored user-mode APIs.
  • Userland Hooking Redirecting an API by patching its code or pointer in user space.
  • DLL Injection Forcing a target process to load an attacker-supplied library.
  • Anti-Debugging Tricks malware uses to detect or resist analysis under a debugger.
  • Anti-VM Checks that detect a virtual machine and suppress malicious behavior.
  • Sandbox Evasion Tactics that detect or stall automated analysis to avoid detonation.
  • Dropper Malware that carries and deposits an embedded payload on the host.
  • Loader Code that prepares and executes a payload, often only in memory.
  • Stager A tiny first stage that pulls and runs a larger second stage.
  • Remote Access Trojan Malware giving an operator interactive remote control of a host.
  • Ransomware Malware that encrypts data and demands payment to restore it.
  • Rootkit Concept Subverting the OS to hide processes, files, or network activity.
  • Fileless Malware Malicious activity that runs in memory without a payload on disk.
  • Living-off-the-Land Binary A trusted built-in binary abused to carry out attacker actions.
  • C2 Configuration Extraction Recovering a sample's embedded settings, like C2 and keys.
  • Byte Signature A fixed byte sequence used to match a known sample on disk.
  • Behavioral Signature Detecting malware by what it does, not how its bytes look.
  • Supply-Chain Implant Malicious code slipped into trusted software before it ships.
  • Malware Family A group of related samples sharing common code or design.
  • Persistence Mechanism The means malware uses to survive reboots and keep running.
  • TTP Mapping Translating observed malware behavior into named adversary techniques.
  • Worm Malware that self-propagates across systems without needing a human to run it.
  • Wiper Malware built to destroy data or render systems unusable, not for profit.
  • Keylogger Malware or hardware that records keystrokes to capture credentials and input.
  • Shai-Hulud A self-replicating worm that spreads through compromised npm packages.

Network Forensics & C2

  • Beaconing Malware 'phoning home' on a regular interval.
  • Command and Control The channel an attacker uses to control compromised hosts.
  • DNS Tunneling Smuggling data or C2 inside DNS queries and responses.
  • Domain Generation Algorithm Malware that algorithmically generates many C2 domains.
  • Data Exfiltration Stealing data out of a network to attacker-controlled systems.
  • Packet Capture Recording raw network frames off the wire for later analysis.
  • Flow Data Summarized session metadata: who talked to whom, without payloads.
  • Deep Packet Inspection Inspecting a packet's payload, past the headers, to identify content.
  • JA3 Fingerprinting Hashing TLS handshake fields to fingerprint the client software.
  • JA4 Fingerprinting A newer, more resilient family of network traffic fingerprints.
  • SNI Inspection Reading the destination hostname from the TLS handshake.
  • Encrypted Traffic Analysis Inferring intent from encrypted traffic's shape and metadata.
  • Domain Fronting Hiding the true destination behind a trusted shared front domain.
  • Fast Flux Rapidly rotating DNS records to keep malicious infrastructure alive.
  • DNS over HTTPS Encrypting DNS queries inside HTTPS or TLS to hide them.
  • DNS Sinkhole Redirecting malicious domain lookups to a controlled server.
  • Port Scanning Probing hosts to discover open ports and reachable services.
  • ARP Spoofing Forging ARP replies to intercept traffic on a local network.
  • Man-in-the-Middle Secretly relaying and possibly altering traffic between two parties.
  • SSL Stripping Downgrading a victim's HTTPS connection to plaintext HTTP.
  • Protocol Tunneling Smuggling one protocol's traffic inside another to evade controls.
  • Covert Channel Communicating through a path not meant to carry data.
  • Data Staging Collecting and bundling stolen data before it leaves the network.
  • Malleable C2 Profile A config that reshapes C2 traffic to mimic benign protocols.
  • Beacon Jitter Randomizing call-home intervals to obscure a beacon's rhythm.
  • Beacon Analysis Finding C2 callbacks by their periodic, repetitive connection pattern.
  • Dead Drop Resolver Hiding the real C2 address in content on a legitimate web service.
  • Peer-to-Peer C2 A decentralized C2 where infected hosts relay commands to each other.
  • Redirector An intermediary host that relays C2 traffic to hide the real server.
  • Reverse Shell A compromised host that dials out to give the attacker a shell.
  • Bind Shell A shell that listens on a port for the attacker to connect in.
  • Web Shell A malicious script on a web server giving remote command access.
  • SOCKS Proxy Pivot Tunneling through a foothold to reach an internal network.
  • Port Forwarding Relaying a port through a host to reach otherwise blocked services.
  • Network Tap A hardware device that copies a link's traffic for monitoring.
  • Intrusion Detection System A sensor that watches traffic and alerts on malicious activity.
  • Intrusion Prevention System An inline sensor that detects and actively blocks malicious traffic.
  • Network Segmentation Dividing a network into zones to contain and control traffic.
  • East-West Traffic Traffic moving between internal hosts rather than in and out.
  • Exfiltration over Cloud Storage Stealing data by uploading it to a legitimate cloud service.
  • Exfiltration over DNS Smuggling data out by encoding it into DNS queries.
  • BGP Hijacking Falsely announcing IP routes to divert or intercept traffic.
  • Low and Slow Pacing activity below thresholds to stay under the radar.
  • Web Service C2 Hiding command-and-control inside trusted web services like paste or cloud apps.

Adversary Tactics

Credential & Identity Attacks

Cloud & Container Security

Cryptography & Data Protection

Detection, Hunting & IR

  • Indicator of Compromise Forensic breadcrumbs that a system was compromised.
  • Dwell Time How long an attacker goes undetected in a network.
  • Chain of Custody The documented trail proving evidence integrity.
  • Threat Hunting Proactively searching for threats that evaded detection.
  • YARA A pattern-matching rule language to classify malware.
  • SIEM Central platform that aggregates logs to alert, search, and investigate.
  • EDR Endpoint agent that records host activity and supports response actions.
  • XDR Detection and response correlated across endpoint, network, and cloud.
  • Detection Engineering The discipline of designing, building, and maintaining detections.
  • Detection as Code Managing detections in version control with testing and CI like software.
  • Alert A notification raised when a detection's conditions are met.
  • False Positive An alert that flags benign activity as malicious.
  • False Negative Malicious activity that no detection caught.
  • Alert Fatigue Desensitization to alerts caused by overwhelming volume and noise.
  • Triage Quickly assessing alerts to decide what's real and what's urgent.
  • Enrichment Adding context to an alert to speed up and sharpen decisions.
  • Correlation Combining multiple events into a single, higher-confidence signal.
  • Detection Rule Logic that defines the conditions under which an alert fires.
  • Signature-Based Detection Detecting known-bad by matching specific, predefined patterns.
  • Anomaly Detection Flagging activity that deviates from an established normal baseline.
  • Behavioral Detection Detecting threats by the actions they take, not static indicators.
  • Baselining Establishing what 'normal' looks like so deviations stand out.
  • UEBA Analytics that profile users and entities to flag risky deviations.
  • Precision and Recall Metrics for how accurate and how complete a detection is.
  • Tuning Adjusting detection logic to cut noise without losing real catches.
  • Hunt Hypothesis A testable assumption about adversary activity that guides a hunt.
  • Crown Jewel Analysis Identifying an org's most critical assets to focus defense on them.
  • Playbook A documented, repeatable procedure for handling a scenario.
  • SOAR Automating and orchestrating response steps across security tools.
  • Escalation Routing an incident to higher tiers or authority as it grows.
  • Incident Severity A classification of an incident's seriousness that drives response.
  • Scoping Determining the full extent of an incident: every affected asset.
  • Containment Stopping an incident from spreading while you investigate.
  • Eradication Removing the threat and the attacker's footholds from the environment.
  • Recovery Restoring systems to normal, trusted operation after an incident.
  • Post-Incident Review A blameless review after an incident to learn and improve.
  • Root Cause Analysis Finding the underlying reason an incident was able to happen.
  • Mean Time to Detect Average time from an intrusion starting to its detection.
  • Mean Time to Respond Average time from detecting an incident to responding to it.
  • On-Call A rotation ensuring someone is always available to respond.
  • Incident Commander The single person who directs and owns an incident response.
  • Purple Teaming Offense and defense collaborating to test and improve detections.
  • Detection Coverage How much of relevant adversary behavior your detections actually catch.
  • Telemetry Gap A blind spot where the data needed to detect activity isn't collected.
  • Honeytoken Fake data planted to fire an alert the moment it is touched.
  • Red Team An authorized team that emulates real adversary tactics to test detection and response.
  • Tabletop Exercise A discussion-based drill that walks a team through a simulated incident scenario.
  • Sigma Rule A vendor-agnostic rule format for writing log-based detection logic.
  • Insider Threat Risk of harm from someone with legitimate authorized access to systems or data.
  • Honeypot A decoy system built to lure, detect, and study attacker activity.

Threat Intelligence & Frameworks

Vulnerabilities & Exploitation

Web, Email & Application Security

Logging & Telemetry

  • Sysmon Free Windows driver that logs rich process, network, and file telemetry.
  • Event Tracing for Windows The high-speed kernel and user-mode tracing framework underpinning Windows telemetry.
  • PowerShell Logging Recording PowerShell commands, scripts, and pipeline activity for audit.
  • Process Creation Auditing Logging every new process start, ideally with its full command line.
  • Antimalware Scan Interface A Windows API that lets security tools inspect scripts and content at runtime.
  • Windows Event Forwarding Native Windows mechanism to ship event logs to a central collector.
  • auditd The Linux kernel audit subsystem's user-space logging daemon.
  • syslog The long-standing standard for emitting and forwarding system log messages.
  • journald systemd's structured, indexed logging service for modern Linux.
  • eBPF Telemetry Safely running sandboxed programs in the Linux kernel to observe activity.
  • Endpoint Detection Telemetry Continuous, detailed activity data collected by endpoint sensors.
  • DNS Logging Recording the domain-name lookups that hosts request and resolve.
  • Proxy Logs Records of outbound web requests passing through a forward proxy.
  • Firewall Logs Records of network connections a firewall allowed or denied.
  • Flow Logs Metadata records summarizing network conversations, without packet payloads.
  • Web Access Logs Server-side records of every HTTP request a web server handles.
  • Authentication Logs Records of login attempts, successes, and failures across systems.
  • Cloud Audit Logs Records of management-plane API activity in a cloud environment.
  • Log Normalization Mapping diverse log formats onto a single common schema.
  • Log Enrichment Adding external context to log events to make them more useful.
  • Common Event Format A standardized, key-value log format for interoperable event sharing.
  • Log Aggregation Collecting logs from many sources into a central, searchable store.
  • Log Retention How long logs are kept available before deletion or archival.
  • Log Forwarding Transmitting logs off their source host to a remote collector.
  • Time Synchronization Keeping system clocks aligned so log timestamps can be correlated.
  • Tamper-Evident Logging Protecting logs so any alteration or deletion is detectable.
  • Structured Logging Emitting logs as well-defined fields rather than free-form text.

Linux, macOS & Mobile Forensics

  • Shell History Files Per-user logs of typed shell commands; a key record of operator activity on Unix.
  • auth.log / secure Linux authentication log recording logins, sudo use, SSH activity, and PAM events.
  • wtmp / btmp / lastlog Binary Unix login-accounting databases behind the last, lastb, and lastlog commands.
  • systemd Journal Structured, indexed binary log managed by systemd-journald on modern Linux.
  • cron and systemd Timers Time-based job schedulers on Unix and a classic persistence and execution vector.
  • /etc/passwd and /etc/shadow Unix account and password-hash files; a focus for backdoor accounts and cracking.
  • sudoers Configuration Policy file granting privileged command execution; a prime privilege-escalation target.
  • SUID / SGID Binaries Executables that run with the file owner's or group's privileges regardless of caller.
  • LD_PRELOAD Hijacking Forcing a malicious shared library to load first so it overrides legitimate functions.
  • Loadable Kernel Module Rootkit Malicious kernel modules that hide artifacts and hook the system from ring 0.
  • /proc Filesystem Virtual filesystem exposing live process and kernel state on Linux.
  • ext4 / XFS Timestamps The atime/mtime/ctime (and ext4 crtime) fields that anchor a Linux file timeline.
  • Anti-Forensic Log Tampering Deleting, editing, or zeroing host logs and history to hide intruder activity.
  • macOS Unified Logging Apple's consolidated, compressed system log spanning macOS and iOS.
  • FSEvents macOS filesystem-change log that records directory-level activity over time.
  • KnowledgeC Database Apple SQLite store of user-activity events used to reconstruct device usage.
  • Gatekeeper Quarantine macOS provenance flag on downloaded files that records source and triggers checks.
  • LaunchAgents and LaunchDaemons launchd jobs that start programs at boot or login, and macOS's top persistence path.
  • TCC Database macOS privacy database recording which apps were granted sensitive access.
  • Credential Store / Keychain OS-managed encrypted vaults holding passwords, keys, and certificates.
  • APFS Snapshots Read-only point-in-time copies of an APFS volume usable to recover prior state.
  • Mobile Acquisition Methods The logical, file-system, and physical tiers of getting data off a phone.
  • iOS Backup Artifacts The structured backup an iPhone produces, a rich offline source of app and system data.
  • SQLite App Databases The SQLite files mobile and desktop apps use to store messages, history, and state.
  • APK Package Android's app-distribution archive; the unit analysts reverse to assess mobile malware.
  • Android Logcat Android's runtime logging stream capturing system and application messages.
  • Mobile Keystore / Secure Enclave Hardware-isolated subsystems that store and use cryptographic keys off the main CPU.
  • Mobile Full-Disk Encryption Always-on encryption that ties phone data to user credentials and device hardware.
  • BFU vs AFU State Whether a phone was unlocked since boot, which governs how much data is decryptable.
  • Jailbreak / Root Removing a phone's OS restrictions to gain privileged, system-wide access.

AI & Automation Security

  • Large Language Model A model trained on large amounts of text that predicts the next token, which lets it write, summarize and answer.
  • Prompt Injection Text crafted to make a language model ignore its instructions and follow the attacker's instead.
  • Indirect Prompt Injection Instructions hidden in content a model reads on someone else's behalf, such as a web page, an email or a document.
  • Jailbreak A prompt that gets a model to produce output its safety training is meant to refuse.
  • System Prompt Leakage Getting a model application to reveal the hidden instructions its developer gave it.
  • Insecure Output Handling Passing a model's output straight into a browser, shell, query or API without checking it.
  • AI Agent A language model given tools and allowed to decide, step by step, which ones to use to reach a goal.
  • Excessive Agency An AI application that can do more than its job needs, so a manipulated model can do real damage.
  • Agent Tool Poisoning A malicious or tampered tool whose description or output steers the agent that uses it.
  • Retrieval-Augmented Generation Fetching relevant documents at question time and handing them to a model so it answers from them.
  • Data Poisoning Tampering with the data a model learns from so that it learns the wrong thing.
  • Model Backdoor A hidden behavior planted in a model that activates only when a specific trigger appears.
  • Adversarial Example An input changed just enough to make a model get it wrong, often in ways a person would not notice.
  • Evasion Attack Using adversarial inputs at run time to slip past a deployed model.
  • Membership Inference Working out whether a specific record was in a model's training data.
  • Model Inversion Reconstructing sensitive training data from a model's outputs.
  • Model Extraction Copying a model's behavior by querying it and training a look-alike on the answers.
  • AI Supply Chain Compromise Tampering with the models, datasets, libraries or tools an AI system is built from.
  • Unsafe Model Deserialization A model file format that can run code when it is loaded, turning a download into an execution.
  • AI Hallucination Confident, plausible output from a model that is simply not true.
  • Guardrails Checks around a model that filter what goes in, what comes out, and what it is allowed to do.
  • AI Red Teaming Attacking an AI system on purpose, before someone else does, to find how it fails.
  • MITRE ATLAS MITRE's ATT&CK-style knowledge base of tactics and techniques used against AI systems.
  • NIST AI Risk Management Framework NIST's voluntary framework for managing AI risk, organized as Govern, Map, Measure and Manage.
  • Deepfake Audio, video or images generated or altered by AI to show someone saying or doing something they did not.
  • Human in the Loop A design where a person reviews or approves an automated decision before it takes effect.
  • Automated Response Containment and cleanup actions that run on their own when a detection meets set conditions.

Related

All references