Back to vendors
C

Cursor

Also known as: Cursor IDE, Cursor Agents, Bugbot

Visit site
Entry priceFree plan, Plus at $8 a month with Grok models only, Premium at $30 a month, Super at $100, Ultra at $200, Teams from $30 per user a month, and Enterprise at custom pricing.Full pricing detail

AI native editor and agent platform built on VS Code, with Tab predictive editing, an agent that runs on Cursor's own Grok and Composer models or outside frontier models, an Agents Window for directing fleets of background and cloud agents, and a plugin marketplace.

Cursor is an AI native code editor built by Anysphere, a San Francisco company founded in 2022 by four MIT students. It is a fork of Visual Studio Code, so existing extensions, themes, and keybindings carry over, and Cursor rebuilds the editor so that AI sits at the center of how code gets written.

The editor has three core capabilities. Tab is a predictive autocomplete that forecasts a developer's next edit, which can span several files. Inline editing lets a developer select code, describe a change in plain language, and have Cursor rewrite it in place. The agent takes a high level instruction, such as adding authentication across a set of endpoints, then plans the work, searches the codebase with its own grep engine and parallel Explore subagents, creates and edits files across the project, runs terminal commands, and iterates until the task is done.

Cursor 2.0 added running several agents in parallel using git worktrees, and Cursor 3 reorganized the product around an Agents Window where directing, monitoring, and reviewing fleets of background and cloud agents is the main workflow. Cloud agents can be started by automations on a schedule or on events such as a new pull request, a Slack message or an incident, and teams can run their tool execution on machines they manage. Cursor Projects, in beta, has a coordinator agent plan larger work such as a migration and delegate it to other agents.

Cursor's own models, Grok and Composer, come with every paid plan, and models from Anthropic, OpenAI and Google run at API list price on Premium and above. Cursor supports MCP for connecting tools and data, controls a built in browser to test applications, and reviews pull requests for bugs with Bugbot. A Privacy Mode keeps code off Anysphere's servers, and enterprises can enforce it across the organization.

SpaceX announced an all stock acquisition of Anysphere in June 2026, and Cursor continues to ship under its own brand. Since October 2026 it is sold as part of The X Subscription, alongside Grok and X Premium.

From our comparison video

Cognition vs Cursor (2026): Which Should You Choose?

Our 4 minute video puts Cursor side by side with Cognition. It covers what each one does and how each is priced, then when to choose which.

Recorded in October 2026. The profile here is live and updates as research is verified.

Vendor details

Canonical URL

https://cursor.com/

Category

Coding agent

Funding status

Built by Anysphere, founded 2022 in San Francisco by four MIT students. In June 2026 SpaceX announced a $60B all-stock acquisition of Anysphere, reported completed in August 2026, placing Cursor under SpaceX's AI unit. Anysphere itself acquired Continue in the same window, so Continue's codebase now sits under the same chain. Since October 2026 Cursor is sold as part of The X Subscription through X Subscription Corp, alongside Grok and X Premium, and continues to ship under its own brand.

Company status

acquired

Use cases & customers

Target customers

developers

Deployment options

Desktop editor (macOS, Linux, Windows)JetBrains tool windowCLI (npm, headless for CI)Mobile appCloud background agents (sandboxed VMs on AWS)

Integrations

Cursor supports MCP servers with enterprise allowlist and blocklist controls. A GitHub App gives background agents repository read and write access, Slack integration lets people address agents conversationally, and the Cursor Marketplace offers plugins bundling skills, subagents, MCP servers, hooks and rules from partners including Amplitude, AWS, Figma, Linear and Stripe. The agent runs terminal commands and edits across the project, and a CLI with a headless flag drops Cursor agents into CI pipelines.

In practice

You need to add authentication across a dozen endpoints and don't want to touch each file by hand. Cursor's agent plans the change, edits the files across your project, runs the commands, and iterates until it works.

You're in the middle of typing and Cursor already knows where you're going next. Its Tab autocomplete predicts your next edit across multiple files from intent, so you tab through changes instead of typing them out.

You have a refactoring sprint and one developer. In Cursor 3's Agents Window, you spin up several background agents on isolated branches in parallel, supervise from one console, and merge the branches that succeed.

Agentic Index coverage score

12.5 / 14 capabilities · 89%

Integrations & Tool Calling Full

MCP servers connect the agent to external tools and data over stdio, SSE or Streamable HTTP with OAuth for remote servers, installed one click from the Cursor Marketplace or configured in mcp.json, and team admins can distribute servers through a team marketplace; tools, prompts, resources and elicitation are supported.

Sourcecursor.com/docs/mcpread 2026-09-29

Workflow Orchestration Full

The agent delegates to subagents that each run in their own context window, in the foreground or in the background and in parallel, with built in Explore, Bash and Browser subagents and custom subagents defined with their own prompts, tool access and models, in the editor, the CLI and Cloud Agents.

Sourcecursor.com/docs/subagentsread 2026-09-29

Knowledge Grounding & RAG Partial

The agent finds code with Instant Grep, an exact match and regex search engine that builds and queries its index on the developer's machine, and an Explore subagent that runs parallel searches per task; files, folders and past chats are attached per conversation with @ mentions. Cursor does not store embeddings of the codebase for search, so context is gathered on each run, and the local grep index exists to speed up search.

Sourcecursor.com/docs/agent/tools/search and help/customization/contextread 2026-09-29

Human Oversight & Guardrails Full

Run Modes decide when the agent stops for approval on shell, MCP and fetch calls: Allowlist runs only listed actions without asking, Auto-review sandboxes shell commands and sends other calls to a classifier that can ask the user to approve, plain English block instructions route named actions to approval, and a team configuration set in the dashboard overrides user and project settings; browser tools require approval by default.

Sourcecursor.com/docs/agent/security/run-modes and docs/agent/tools/browserread 2026-09-29

Security, Identity & Governance Full

Cursor holds a SOC 2 Type II attestation and AIUC-1, ISO/IEC 27001:2022 and ISO/IEC 42001:2023 certifications, with certificates and reports available on request at trust.cursor.com, and offers SSO, SCIM and enforced multi factor authentication. SAML and OIDC SSO come on Teams, and SCIM, repository and model access controls and audit logs on Enterprise.

Sourcecursor.com/security and cursor.com/pricingread 2026-09-29

Observability & Auditability Full

Hooks fire on every step of the agent loop (preToolUse and postToolUse, shell and MCP execution before and after, file reads and edits, subagent start and stop, agent responses and thoughts), auditing is one of their named uses, and Enterprise admins distribute team hooks from the dashboard.

The Admin API returns per request usage events with model, tokens and conversation ID, and its audit log covers logins and user and settings changes, with no agent actions among them. Cursor does not ship a log sink for the per step hook record, so the customer builds one to keep that record as an audit trail.

Sourcecursor.com/docs/hooks and docs/account/teams/admin-apiread 2026-09-29

Memory & State Persistence Partial

Past agent transcripts can be searched from the Agents Window and attached to a new conversation with @Chats, and checkpoints snapshot modified files within a session. Cursor does not describe a memory layer of its own, project rules act as standing instructions, and subagents start without prior conversation history.

Sourcecursor.com/help/customization/context and docs/agent/overviewread 2026-09-29

Deployment & Data Residency Full

Self-Hosted Machines move Cloud Agent tool execution to machines the customer manages, such as a Linux VM in the customer's AWS account, a Kubernetes cluster or its own devices, with the checkout, build cache and machine local credentials staying there (Team Pools on Enterprise). Cursor still runs the agent loop, inference and planning, and Enterprise customers can enroll in US only data residency.

Sourcecursor.com/docs/cloud-agent/self-hosted and help/security-and-privacy/regionsread 2026-09-29

Prebuilt Agents, Templates & Packs Full

Plugins bundle rules, skills, agents, commands, MCP servers and hooks, and the Cursor Marketplace lists manually reviewed official plugins installed per project or user from Customize, including Amplitude, Datadog, Figma, Linear, Sentry, Stripe, Supabase and Vercel plugins. Teams and Enterprise can run private team marketplaces, and three subagents ship built in.

Sourcecursor.com/docs/plugins and cursor.com/marketplaceread 2026-09-29

Triggers & Channel Coverage Full

Automations run cloud agents in the background on a schedule (preset or cron) or in response to events, a GitHub pull request opened or marked ready, Slack messages, Linear issue changes, PagerDuty incidents, Sentry errors and webhooks, with no person starting the run; Bugbot also reviews every pull request update automatically.

Sourcecursor.com/docs/cloud-agent/automations and docs/bugbotread 2026-09-29

Model Flexibility & Routing Full

Cursor's own models, Grok and Composer, run from a weekly usage pool on every paid plan. On Premium and above, users also pick Anthropic, OpenAI and Google models at API list price from prepaid credit. On Teams and Enterprise, Cursor Router picks among Cursor and third party models by cost, balance or intelligence, and admins allow or block models. Plus is limited to Grok models.

Sourcecursor.com/docs/models-and-pricing and cursor.com/pricingread 2026-10-25

APIs, SDKs & MCP Extensibility Full

The Cloud Agents API at api.cursor.com, with Basic or Bearer authentication on user or service account keys, creates, lists, follows up, streams, cancels, archives and deletes agents and their runs and returns artifacts, and TypeScript and Python SDKs are published beside it.

Sourcecursor.com/docs/cloud-agent/api/endpointsread 2026-09-29

Testing, Debugging & Optimization Partial

Bugbot reviews each pull request for bugs, security issues and quality problems and publishes a CI check status, and a custom verifier subagent can run the tests and report what passed. Both test the customer's code, and Cursor does not describe a harness, scored test set or optimization loop for evaluating the agent's own behavior.

Sourcecursor.com/docs/bugbot and docs/subagentsread 2026-09-29

Browser & Computer Use Full

The browser tool lets the agent control a browser pane inside Cursor, navigating, clicking, typing, scrolling and taking screenshots to test applications, audit accessibility and check visual changes, through Cursor's own MCP server, with approval required by default and an admin origin allowlist; a built in Browser subagent runs these steps, and self hosted workers run computer use tools.

Sourcecursor.com/docs/agent/tools/browser and docs/subagentsread 2026-09-29

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Product updates

·Agent capabilityVerified

Realtime voice calls with agents now work in the Cursor IDE and on the web, alongside the Agents Window and the iOS app. A voice agent running on Grok Voice takes spoken requests, hands coding work to the chat agent and speaks short progress updates while that agent runs. With Privacy Mode on, calls run with zero data retention.

Bears on: Agent capability

View source
·Agent capabilityVerified

Cloud agents in Cursor can now build and publish live web apps. The agent scaffolds the app, sets up a Postgres database when needed, puts sign in behind Cursor SSO and hosts it on a domain the user picks, and nothing goes live until deterministic and agentic security checks pass and the user explicitly publishes. Team admins can see every app and secret, add their own review rules, approve blocked publishes and unpublish any app.

Bears on: Deployment / data residency

View source
·Human approval / guardrailsPartially Verified

Cursor 3.24 makes Auto-review stop for approval before the agent runs a recursive, force, wildcard or mirror delete whose target could resolve differently than written, and commands that match a user's block instructions now always show the approval card. A new managed admin policy, AgentRemoteWorkspaceAuthorities, turns Agent off in local Cursor windows and allows it only in approved remote workspaces.

Bears on: Security / enterprise

View source
View all 14 Cursor updates →Tracked since May 2026 · Verified from public vendor sources

Pricing

Free plan, Plus at $8 a month with Grok models only, Premium at $30 a month, Super at $100, Ultra at $200, Teams from $30 per user a month, and Enterprise at custom pricing.

A monthly plan with a weekly usage pool for Cursor's own models, Grok and Composer. Models from other makers are paid from prepaid credit at API list price on Premium and above.

Free tier

Included quota

Each paid plan shares one weekly usage pool for Cursor models (Grok and Composer). Premium carries four times the usage of Plus, Super 15 times and Ultra 50 times. Absolute pool sizes are not published, and third party models are not included in any plan.

What is public

Since October 2026 Cursor is sold as part of The X Subscription, through X Subscription Corp, which also carries Grok and X Premium benefits. Free has limited agent and Chat requests. Plus is $8 a month with Grok models only. Premium is $30 a month and adds the latest Grok models, cloud agents and automations, code hosting, Bugbot code reviews and Grok Bot. It also opens models from other makers at API list price. Super is $100 and Ultra $200, with 15 and 50 times the usage of Plus. Teams starts at $30 per user a month, with seats at $30, $100 or $200, centralized billing, usage analytics, team wide privacy settings, SAML and OIDC SSO and a team marketplace. Enterprise is custom, adding invoice or PO billing, SCIM, granular spend controls, audit logs, service accounts and priority support. Legacy plans (Pro $20, Pro+ $60, Ultra $200, Teams $40 and $120 per user) continue for existing subscribers unless they switch.

Billing mechanics

Plans are billed monthly per account, or per user on Teams, and each carries a weekly pool of included usage for Cursor's own models. Usage beyond the pool, and all third party model usage, is paid from prepaid credit, with optional auto top up. Enterprise pools usage across seats and bills by invoice or PO. Legacy plans keep on demand usage in place of top ups.

Cost watchouts

Real cost depends on model choice and agent volume beyond each plan's included usage. Outside models are always paid from prepaid credit at API list price, and Teams and Enterprise add $0.25 per million tokens on eligible third party requests.

Variable cost rationale

Cursor meters model usage. Each paid plan includes a weekly pool for its own Grok and Composer models, and agent runs draw it down at different rates by model. Work beyond the pool, and every request to a Claude, GPT or Gemini model, is paid from prepaid credit at API list price, so the bill grows with how much agent work runs on outside models. The flat plan price and the weekly pool are the main limits, and Super and Ultra carry 15 and 50 times the usage of Plus. A monthly additional usage limit caps paid usage regardless of the credit balance, and Teams share one prepaid balance and cap. Teams and Enterprise pay an extra $0.25 per million tokens on eligible third party requests.

Additional watchouts

Heavy agent work beyond the weekly pool, and any use of outside models, draws prepaid credit, so the plan price is a floor rather than a total.

Overage / add-ons

Cursor models beyond the weekly pool, and all third party model usage, are paid from prepaid credit at per token rates, with auto top up and a monthly additional usage limit. On Teams and Enterprise, eligible third party requests add a Cursor Token Rate of $0.25 per million tokens.

Sales call required

Mixed (some tiers require a call)

Free / trial

Free plan with limited agent and Chat requests.

Lowest paid plan

Plus at $8 per month, with Grok models only

Commercial notes

Free, Plus, Premium, Super, Ultra and Teams are bought self serve. Enterprise is sold through sales with invoice or PO billing.

Key ambiguities

Usage pools are stated as multiples of Plus, not as absolute amounts.

Cancellation / refund

Hobby needs no credit card, Individual and Teams are self serve, and Enterprise is custom with invoice or PO billing. Annual terms, trials and student offers are not published.

Support SLA / resale

Teams adds centralized billing, administration and usage analytics, and Enterprise adds priority support.

Missing data

The size of the weekly usage pool, all Teams seat prices, and Enterprise pricing are not public.

Agentic Index verified 2026-10-25

Alternatives to Cursor

The closest documented capability profiles to Cursor among coding agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.