Reference
AI coding & terminal glossary
The words you keep hearing in AI-first development, explained plainly. Every entry links to deeper docs or articles where they make a difference.
A
- Agentic codingAgentic coding is software development where an LLM-powered agent plans, edits, runs, and verifies code on its own using tools, not just autocomplete.
- AI pair programmingAI pair programming is a collaboration style where an LLM assistant sits alongside you, suggesting code and reviewing changes in real time as you work.
- ANSI escape codesANSI escape codes are control sequences that terminals interpret for colors, cursor movement, and screen clearing — the language of every modern CLI UI.
- Autonomous agentAn autonomous agent is an AI program that perceives, decides, and acts on its own toward a goal — the architecture behind modern coding CLIs.
C
- CheckpointA checkpoint is a saved snapshot of file state that lets you roll back an AI coding agent's changes to a known-good point.
- Claude CodeClaude Code is Anthropic's official command-line agent that plans, edits, runs, and verifies code across your repo using Claude models and tool use.
- CLI wrapperA CLI wrapper is a program that launches, manages, and UI-wraps other command-line tools — how apps like SpaceSpider host Claude Code and Codex CLI.
- Codex CLICodex CLI is OpenAI's open-source command-line coding agent that runs GPT models against your local repo with tool use, sandboxing, and approvals.
- ConPTYConPTY is the Windows pseudo-console API that lets apps host real terminal sessions with proper ANSI, resize, and signal support.
- Context windowThe context window is the maximum number of tokens an LLM can consider at once — the hard limit on how much conversation and code it can see.
E
F
G
- Git worktreeGit worktrees let you check out multiple branches of the same repo into separate directories simultaneously — ideal for parallel AI coding agents.
- Grid layoutA grid layout arranges multiple terminal panes in rows and columns so you can watch several CLIs at once — the core UI pattern for AI-first terminals.
H
K
L
M
P
- Plan modePlan mode is a read-only Claude Code state where the agent investigates and proposes a plan before making any file edits or running commands.
- Prompt engineeringPrompt engineering is the practice of writing inputs that steer an LLM toward useful, accurate, on-task output — the core skill of every AI power user.
- PTYA PTY is a virtual terminal pair that lets programs talk to each other as if through a physical serial terminal, enabling real shells inside apps.
Q
R
S
- SandboxA sandbox is an execution boundary that limits what an AI coding agent can touch — which files, processes, or network resources it can access.
- ShellA shell is the command interpreter that runs inside a terminal, parses commands, manages jobs, and provides scripting — bash, zsh, fish, and PowerShell are the popular ones.
- Slash commandA slash command is a reusable named prompt macro invoked with a leading slash, like /review or /test, used to trigger repeatable agent workflows.
- SpaceA space in SpaceSpider is a saved workspace: a directory plus a grid of terminal panes, each running a chosen CLI like Claude Code or a shell.
- SubagentA subagent is a specialized LLM agent spawned by a parent agent to handle a focused subtask with its own prompt, tools, and context window.
- System promptA system prompt is the persistent instruction block that sets an LLM's role, rules, and tools for the entire conversation.
T
- Terminal multiplexerA terminal multiplexer runs multiple shell sessions inside one terminal window, with splits, tabs, and detachable sessions. tmux and Zellij are the popular ones.
- tmuxtmux is a terminal multiplexer that runs multiple shells in one window with splits, tabs, and detachable sessions that survive disconnects.
- TokenA token is the atomic unit an LLM processes — typically a short piece of text (a word, part of a word, or symbol) produced by a tokenizer.
- Tool useTool use is the mechanism by which an LLM requests an external action — like reading a file or running a shell command — via structured function calls.
- TTYTTY is the Unix abstraction for a terminal device, inherited from teletypewriters. It defines stdin, stdout, signals, and line discipline for every CLI.
V
- Vector databaseA vector database stores and searches high-dimensional embeddings using similarity metrics, powering semantic search, RAG, and recommendation.
- Vibe codingVibe coding is an informal style of building software where you describe intent to an AI agent and accept its output without reading every line.
W
- WarpWarp is a GPU-accelerated terminal with block-based commands, AI features, and team sharing — a modern take on the traditional terminal emulator.
- Windows TerminalWindows Terminal is Microsoft's modern tabbed terminal app for Windows, built on ConPTY with support for PowerShell, cmd, WSL, and custom profiles.