Cursor
Also known as: Cursor IDE, Cursor Agents
AI-first IDE with built-in agentic coding assistance, multi-file editing, code review, and frontier model support.
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, but instead of bolting AI on as a plugin, Cursor rebuilds the editor so that AI sits at the center of how code gets written. It has become one of the most widely used tools in professional software development and is adopted across much of the Fortune 500.
Three capabilities anchor the experience. Tab is a predictive autocomplete that forecasts a developer's next edit, sometimes across several files, rather than just finishing the current line. Inline editing lets a developer select code, describe a change in plain language, and have Cursor rewrite it in place. And the Composer agent takes a high-level instruction, such as adding authentication across a set of endpoints, then plans the work, creates and edits files across the project, runs terminal commands, and iterates until the task is done, using a semantic index of the whole codebase to understand dependencies.
Through 2025 and 2026 Cursor moved steadily toward autonomous, multi-agent work. Cursor 2.0 added running several agents in parallel using git worktrees, and Cursor 3, released in April 2026, reorganized the product around an Agents Window where directing, monitoring, and reviewing fleets of background and cloud agents becomes the primary workflow, with the text editor stepping back. Background and cloud agents can keep working in isolated sandboxes while a developer does other things, and sessions can move between local and cloud.
Cursor is model-flexible, routing between frontier models from Anthropic, OpenAI, Google, and xAI, while also developing its own in-house Composer models tuned for fast, cost-effective agentic coding. It supports MCP for connecting tools and data, includes an automated bug-finder for pull requests, and offers a Privacy Mode that keeps code off Anysphere's servers, which enterprises can enforce organization-wide.
In June 2026, SpaceX announced it would acquire Anysphere, placing Cursor under its xAI subsidiary, in a deal expected to close in the third quarter of 2026.
Vendor details
Canonical URL
https://cursor.com/
Category
Coding agent
Company status
unclear
Use cases & customers
Target customers
Deployment options
In practice
You need to add authentication across a dozen endpoints and don't want to touch each file by hand. Cursor's Composer agent plans the change, edits the files across your project, runs the commands, and iterates until it works.
You're mid-keystroke 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
11.0 / 14 capabilities · 79%
| Integrations & Tool CallingAgentic AI Index Expansion Evidence Report | Full |
|---|---|
| Workflow OrchestrationAgentic AI Index Expansion Evidence Report | Full |
| Knowledge Grounding & RAGAgentic AI Index Expansion Evidence Report | Partial |
| Human Oversight & GuardrailsAgentic AI Index Expansion Evidence Report | Partial |
| Security, Identity & GovernanceAgentic AI Index Expansion Evidence Report | Full |
| Observability & AuditabilityAgentic AI Index Expansion Evidence Report | Full |
| Memory & State PersistenceAgentic AI Index Expansion Evidence Report | Partial |
| Deployment & Data ResidencyAgentic AI Index Expansion Evidence Report | Partial |
| Prebuilt Agents, Templates & PacksAgentic AI Index Expansion Evidence Report | Partial |
| Triggers & Channel CoverageAgentic AI Index Expansion Evidence Report | Full |
| Model Flexibility & RoutingAgentic AI Index Expansion Evidence Report | Full |
| APIs, SDKs & MCP ExtensibilityAgentic AI Index Expansion Evidence Report | Full |
| Testing, Debugging & OptimizationAgentic AI Index Expansion Evidence Report | Partial |
| Browser & Computer UseAgentic AI Index Expansion Evidence Report | Full |
The Agentic Index coverage score grades every vendor Full, Partial or Unable to verify 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
Recent platform changes
Cursor introduced new plugins that allow its coding agents to directly access Google Workspace applications. The integration enables agents to read, write, and act across Gmail, Google Drive, and Google Calendar directly from the IDE.
Bears on: Integrations
View sourceCursor has introduced Cursor Router, an intelligent model-routing layer that powers its Auto mode for Teams and Enterprise plans. It automatically analyzes each coding request and routes it to either a frontier model for complex reasoning or a cost-efficient model for routine tasks. The update includes three optimization modes: Intelligence, Balance, and Cost.
Bears on: Agent capability
View sourceCursor released enhancements to its Slack integration, enabling the agent to share an execution plan before beginning a task. The update adds support for multi-repo environments directly from Slack and introduces cross-channel workflows, allowing the agent to pull context from other channels or threads.
Bears on: Integrations
View sourcePricing
From $20/mo · free tier
hybrid
Included quota
Hobby (free): limited Tab completions (~2,000/mo) + ~50 slow premium requests/mo, full editor, codebase indexing, one-week Pro trial. Pro ($20/mo): unlimited Tab, extended agent limits, frontier-model access, MCPs/skills/hooks, cloud (Background) agents, $20 credit pool, unmetered Auto. Pro+ ($60): 3x credits. Ultra ($200): 20x credits + priority feature access. Teams Standard ($40/user, $20 included usage): shared chats/rules, centralized billing, analytics, RBAC, org privacy mode, SAML/OIDC SSO. Teams Premium ($120/user): 5x Standard usage. Enterprise: pooled usage, SCIM, AI-code-tracking API + audit logs, admin/model controls, priority support.
What is public
Cursor (cursor.com, by Anysphere - AI-native code editor with agent mode, cloud agents, MCP, and Composer multi-file editing) has six tiers plus a Bugbot add-on. Hobby: free (permanent, ~2,000 Tab completions + ~50 premium requests/mo, full editor, 7-day Pro trial). Pro: $20/mo ($16 annual) - unlimited Tab, frontier models, cloud agents, a $20 monthly credit pool, and unmetered Auto mode. Pro+: $60/mo (3x credits). Ultra: $200/mo (20x credits). Teams: $40/user/mo Standard ($32 annual; $120 Premium seat = 5x usage, 3x cost). Enterprise: custom. Annual billing saves 20%; students get a free year of Pro.
Billing mechanics
Since a June 2025 overhaul, each paid plan includes a monthly CREDIT POOL equal to its price (Pro = $20 of credits), and billing is tied to underlying model API rates. Crucially, 'Auto' mode (Cursor routes to the best model) is UNLIMITED and does NOT draw from the pool - so everyday coding stays effectively unmetered. Manually selecting a premium model (Claude Opus/Sonnet, GPT, Gemini) or enabling 'Max Mode' (bigger context) deducts from the pool at API cost. Once the pool is spent, Auto keeps working; premium models pause until reset or you enable on-demand usage (billed in arrears, set your own spend cap) or buy more credits. Per-seat licensing on Teams ($40 Standard / $120 Premium per user; the June 2026 split also gives Standard seats two usage pools - first-party Composer/Auto + third-party API).
Cost watchouts
The credit pool drains fast if you manually pick premium models (e.g., Claude Opus), enable Max Mode, or work in large codebases - a few Max-Mode queries can wipe a Pro pool, and heavy agent/Composer use can exhaust it in ~2 weeks; on-demand overage has no default cap (set your own); Teams Standard seats ($40) are 2x Pro and include the same $20 usage as individual Pro; Bugbot (PR review) is a separate add-on with its own tiers (Free/Pro/Teams/Enterprise); annual billing locks you in
Variable cost rationale
Hybrid - a flat per-seat base plus a usage-metered credit pool; everyday Auto-mode work is effectively unlimited (low exposure), but manual premium-model use, Max Mode, large-context work, and continuous cloud agents consume credits at API rates and can trigger on-demand overage
Additional watchouts
Credit-based billing is the main friction - manual premium models + Max Mode + big codebases burn the pool fast and create cost anxiety (the June 2025 rollout drew heavy backlash, with refunds/apology July 4 2025); Teams Standard seats cost 2x Pro but include the same $20 usage; on-demand overage needs a manual spend cap; Bugbot is an extra line item; annual lock-in for the 20% discount
Overage / add-ons
Auto mode is always unlimited (no pool draw). When the monthly credit pool is exhausted, premium-model selection stops until reset - OR you enable on-demand usage (billed in arrears at the model's API rate, with a spend limit you set) or buy extra credits. BYO API key works for some models, but Tab and apply-from-chat always run on Cursor's own models. Enterprise uses pooled credits across seats.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free (Hobby tier implied)
Lowest paid plan
Team $40/user/mo
Commercial notes
One of the most popular AI-native editors (VS Code fork by Anysphere); biggest differentiators vs GitHub Copilot/Windsurf are agent mode, parallel CLOUD agents on remote VMs, Composer multi-file editing, and the mature MCP/plugin ecosystem; pricier than Copilot ($10 Pro / $19 Business) but more agentic; Windsurf ($15 indiv / $30 team) is the cheaper direct competitor; separate Bugbot product for AI PR reviews
Key ambiguities
Real cost depends entirely on model choice and Auto-vs-manual usage (the June 2025 credit shift made the sticker price look smaller for light use and larger for heavy use); the Teams Standard ($40)/Premium ($120) seat split was introduced June 2026 and rolls to existing teams on renewals from July 1, 2026; Enterprise pricing is custom
Cancellation / refund
Hobby is free forever (no card); Pro/Pro+/Ultra are self-serve monthly or annual (20% off annual, locks in); a one-week Pro trial on signup; students get a free year of Pro (.edu + SheerID); Teams self-serve per-seat; Enterprise is custom (invoice/PO billing)
Support SLA / resale
Community/standard support (individual tiers); Teams adds centralized billing, usage analytics, RBAC, org-wide privacy mode, and SAML/OIDC SSO; Enterprise adds pooled usage, SCIM, AI-code-tracking API + audit logs, granular admin/model controls, and priority support/account management; mature one-click MCP marketplace (Figma/Linear/Stripe/Vercel/AWS) and plugin system
Missing data
Real monthly cost is usage-dependent (Auto vs manual model, Max Mode, codebase size) and isn't a single number; Enterprise pricing is custom; the Teams Standard/Premium seat split (introduced June 2026) rolls to existing teams from July 1, 2026. The individual paid floor: Pro $20/mo ($16 annual); Teams $40/user/mo (Standard) with a new $120 Premium seat; Hobby is a permanent free tier.
Related vendors
- Cognition — Maker of Devin, an autonomous AI software engineer
- 10Web — Agentic website platform whose team of ten specialized AI agents…
- Ageiro — Early stage enterprise platform that turns business goals into…
- Aider — Open-source, model-agnostic terminal coding agent that works…
- Anthropic Claude Code — Anthropic's terminal and IDE-native coding agent with sub-agent…
- AppFactor — Agentic orchestration platform that autonomously maintains and…
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.
- OpenHands11.5 / 14Fuller documented coverage on Deployment & Data Residency
- Google Antigravity10.0 / 14Fuller documented coverage on Human Oversight & GuardrailsCursor vs Google Antigravity →
- Factory12.5 / 14Fuller documented coverage on Human Oversight & Guardrails and Deployment & Data ResidencyCursor vs Factory →
- Kilo Code9.5 / 14Fuller documented coverage on Deployment & Data Residency
- Zed9.5 / 14Fuller documented coverage on Deployment & Data ResidencyCursor vs Zed →
- Codegen10.0 / 14Fuller documented coverage on Human Oversight & Guardrails and Deployment & Data Residency
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded