Cursor
ActiveAI-first code editor built on VS Code with deep agent integration.
The most polished AI editor experience today, but usage-based pricing makes costs unpredictable for heavy agent-mode users — budget before you standardize on it.
Agent mode runs inside the loop holding context; tab autocomplete is a stateless per-keystroke call — two first-class modes.
Is it right for you?
Good for
- Teams already on VS Code: extensions, keybindings, and settings carry over
Not good for
- Predictable budgeting — usage-based pricing makes heavy agent months expensive
Our experience
We ran Cursor across our own product work for four months, including agent-mode sessions on multi-file refactors. The editor experience is consistently ahead of plugins-on-VS-Code setups; the pricing change mid-trial moved our team cost up 38% at unchanged usage.
Pricing
Freemium- Limited Agent requests
- Limited Tab completions
- Extended Agent limits
- Frontier model access
- MCPs, skills, hooks
- Cloud agents
- Bugbot on usage-based billing
- 3x usage of Pro
- All Pro features
- 20x usage of Pro
- Priority access to new features
- Centralized billing
- Team marketplace
- Agentic code reviews
- Cloud agents + automations
- Usage analytics
- SAML/OIDC SSO
- Pooled usage
- SCIM seat management
- Repository/model/MCP access controls
- Audit logs
- AI code tracking API
No verdict changes yet
The clock starts day one — changes land here as our verdict evolves.
Sources
- Cursor pricing pageJun 2026
- Cursor changelogJun 2026
Verification log
- Pricing— No changes
Automated agent
- Pricing— Updated
Automated agent
Pricing restructured: old Hobby/Pro(usage-based from $20)/Business(usage+admin) → new Hobby/Pro($20/mo)/Pro+($60/mo)/Ultra($200/mo)/Teams($40-120/user/mo)/Enterprise(Custom). Moved from usage-based to tiered individual plans with Pro+/Ultra tiers.
- Profile— No changes
Imported at launch
- Pricing— No changes
Imported at launch