The file format for a Cursor rule; set always_apply to false and scope it so the rule only fires on the files that need it instead of burning context every request.
A prompt invocation that runs your built app in a browser pane inside Cursor; the agent can then take control of it to click, scroll and test the UI itself.
An open standard that lets any IDE host any coding agent; Cursor uses it to run inside JetBrains/IntelliJ and Android Studio via the Cursor CLI bridge.
Cursor's agent-first window: the agent chat sits center, history on the left and the codebase on the right, with a toggle back to the classic editor. Often called the Agents window.
Cursor's full-capability mode: the AI can read the codebase, write and edit files, move them and run terminal commands. Contrast with Ask mode, which is read-only.
A product spec kept as a file in the repo — covering the PR breakdown, stages and architecture — so the agent can treat it as a source of truth and you can run code review on the spec itself.
The US revenue-recognition standard; cited as the canonical judgment-heavy accounting work to keep human-led rather than hand to an agent, because facts and circumstances vary deal to deal.
Cursor's autoinstall system, in which an earlier Composer model sets up each RL training environment. One agent proposes 10 commands with their expected output, a second agent works to make three of them run, and an environment that still fails after five attempts is discarded.
The Cursor Settings > Agents setting that decides which commands the agent runs automatically and which wait for your approval, via allow and block lists.
Sampling a model N times and keeping the best answer. During Composer 2's RL, best-of-16 on a held-out set and best-of-4 on CursorBench rose along with the average, which Cursor's report takes to suggest RL also widened the set of correct solutions.
Anthropic's previous top-priced Cursor model, hidden by default and succeeded by Claude Fable 5.1; it carries the same data-retention approval requirement.
Anthropic's highest-priced Cursor model for long, complex work at $10 per million input tokens and $50 per million output, with $0.25 cache reads; Enterprise, Teams and Privacy Mode accounts need data-retention approval to use it.
Anthropic's small, low-cost model in Cursor at $0.10 per million input tokens and $0.50 per million output for requests up to 100k input tokens, with 5x rates above that.
Anthropic's medium-tier model in Cursor at $2 per million input tokens and $10 per million output, with cache reads at $0.10; it keeps Claude Sonnet 5's input and output rates and succeeds it.
Agents that run in a Cursor-managed virtual machine, check out the repo, do the work and open a pull request, then shut down, with no load on your laptop.
Cursor's own coding model, built on the open Kimi K2.5 base as a software-engineering specialist. Cursor calls it frontier-level at coding and more cost-efficient than comparable models; Composer 2.5 has since replaced it in Cursor's model list.
The quality drop that happens when one chat accumulates unrelated tasks or tangents, burying the signal the model needs in a haystack of stale context.
Further training of an existing base model on a large, code-heavy data mix before reinforcement learning starts, to deepen its coding knowledge. For Composer 2, Cursor spent most of the compute on 32K-token sequences, extended to 256K, and finished with a short fine-tuning phase on targeted coding tasks.
The security principle, missing in today's LLMs, that a system should distinguish trusted instructions from untrusted input; its absence is why prompt injection works.
A Cursor analytics view that passively categorises what agents are doing (new features, bug fixes, refactors) so leaders can see where engineering time goes.
The longest end-to-end chain of dependent work in a plan; it sets the realistic timeline no matter how parallel everything else looks. A slip on the critical path moves the date; a slip on a task with slack does not.
Cursor's larger-screen mobile app for starting and directing cloud agents, reviewing full pull requests and marking up changes; it is not a full local IDE.
Cursor's hosted layer around each model (context selection, caching, retries) that makes the same model run better and cheaper than calling it directly.
Small scripts Cursor runs at fixed points in an agent's lifecycle (such as before a prompt or before a file write) to enforce policy deterministically.
On legacy Cursor plans, the monthly usage pool for Cursor's own models (Grok 4.7, Grok 4.6, Grok 4.5 and Composer 2.5), kept separate from the third-party Other Models pool. The X Subscription replaces it with one weekly pool.
An India-only legacy individual plan at ₹649 a month centered on Cursor's own models, with cloud and mobile access but without Other Models, Auto, Bugbot, Automations or the Agent SDK; Cursor names Plus as the comparable plan on The X Subscription.
Probing a running application for vulnerabilities from the outside, the way an attacker would. Complements SAST, which reads the code instead of poking the app.
The shift from writing a text spec to building a live, clickable prototype to convey product intent; what gets cheap to make moves the leverage to vision and taste.
The contract spelling out exactly how a vendor may process your data: purposes, subprocessors, retention and breach duties. The document privacy reviews actually read.
Network policy for where Cloud Agents may send traffic: an access mode (Allow all / Default + allowlist / Allowlist only) set per user or per saved environment, with Cursor's own services and SCM providers always reachable; Enterprise admins can lock it org-wide. A denylist with DNS filtering and an HTTP proxy has been described as planned, not shipped.
An engineer who embeds inside a customer's own engineering org to scope ambiguous problems and ship production-grade Cursor workflows a senior engineer keeps using after they leave.
A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product.
Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate.
An early AI teammate product from SpaceXAI and Cursor that works across apps on a persistent cloud computer and asks for approval on selected actions; launch materials describe it as beta, but Cursor says Beta Services terms do not apply.
Cursor as a model-neutral test bed: frontier labs send their newest models to Cursor pre-release to tune them inside the same agent harness every model shares.
Card-processing pricing that passes the underlying network and issuing-bank cost through to you with a fixed markup on top, giving full visibility into cost — unlike a blended sticker rate.
Cursor's built-in browser, driven by Playwright under the hood, so the agent can navigate, click, screenshot and capture each action's metadata while you watch.
Moonshot's open model that Cursor used as the base for Composer 2 and Composer 2.5 (1T parameters, 32B active, 256K context). Cursor's report says it chose Kimi K2.5 for strong results on its internal tests and for its efficiency on Cursor's infrastructure.
Extra Level 2 / Level 3 transaction metadata that can qualify certain US card transactions for cheaper interchange or scheme fees; supplying it correctly is a real cost-optimization lever.
The plans Cursor sold before The X Subscription (Hobby, Start, Pro, Pro+, Ultra, and Teams Standard and Premium seats); they keep working for existing subscribers with two monthly usage pools and on-demand usage.
A penalty on effort added to the reward during RL. In Composer 2's training it grows with effort, but each extra unit costs less than the last, so easy tasks finish quickly while hard tasks can run longer.
Reinforcement learning on realistic coding tasks that take many steps: the model makes a group of attempts at each problem and is updated toward the better solutions. Cursor says Composer 2 can solve tasks that need hundreds of actions, and its Composer 2.5 post says rollouts can span hundreds of thousands of tokens.
The overarching contract between customer and vendor that everything else (order forms, DPAs, SLAs) hangs off. Negotiated once so each later purchase doesn't restart legal review.
The modern single sign-on standard, built as an identity layer on top of OAuth 2.0. Where SAML is XML and enterprise-legacy, OIDC is JSON and what newer tools implement first.
On legacy Cursor plans, the monthly pool for supported third-party models from providers such as Anthropic, Google and OpenAI, separate from Cursor Models. On The X Subscription third-party models are billed from prepaid credit instead.
A Cursor marketplace package that bundles MCP servers and skills (sometimes sub-agents and hooks); one click installs all of it into your Cursor instance.
The spec describing what to build, for whom and why, before implementation starts. Increasingly the artifact you hand an AI agent as the source of truth for a feature.
Balance bought by top-up or Auto Top-up on The X Subscription that pays for third-party models from the first request, and for Cursor models beyond the weekly pool, at the same per-model rates.
Cursor's setting that guarantees code data is not used for training by Cursor or its model providers, and that an admin can enforce org-wide; data-retention terms are a separate, contractual layer.
How skills load: the agent boots knowing only each skill's name and description, then pulls a skill's full instructions into context only at the moment it decides to use it.
A living program-tracking log: Risks (might go wrong), Assumptions (treated as true but unverified), Issues (a risk that already materialized), Dependencies (work blocked on someone else) — each with a single named owner.
The XML-era enterprise standard that powers single sign-on: your identity provider vouches for you to each app. Older than OIDC but still what many enterprise tools speak.
An agent execution mode in which terminal commands run in a restricted environment that blocks unauthorized file access and network activity, with network limited to the sandbox allowlist; admins enable it org-wide through the Run Mode policy.
Scanning source code for vulnerabilities without running it. In AI-heavy workflows it is the automated gate that reads every generated diff before a human does.
A trained behavior in which the model, on reaching a fixed token-length trigger, summarizes its own context and continues from that summary, so one task can run past the context window. Cursor trains Composer this way and rewards the summaries along with the rest of the run.
Short for separation of duties: the rule that authoring, approving and deploying a change never sit with one person. Auditors use the acronym; the control is the same.
The subscription Cursor is sold through: one plan for Grok, Cursor and X, with Free, Plus, Premium, Super and Ultra tiers, per-seat Teams, a weekly included pool for Cursor models shared with Grok, and third-party models billed at API list price from prepaid credit.
A reasoning model (shown with a brain icon in Cursor's picker) that spends extra compute before answering; reach for it on complex, nuanced work and a standard model for fast, simple tasks.
The included usage on a paid plan of The X Subscription: one pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5 and Composer 2.5), shared across Cursor and Grok and refreshed weekly with no rollover.
An MCP server Apple maintains inside Xcode that exposes the simulator, build logs and debugging to Cursor, so the agent can run and iterate on iOS apps it writes.