Consider it for
Developers who want code assistance and agent work inside an editor workflow.
Ai Coding · Tool profile
Cursor
Coding editor and agents for repository work, planning and review.
Plan at a glance
Plans & sources ↓- Starting plan
- Free
- Includes
- See the plan limits below.
Limits & source scope
Model allowances and optional on-demand usage affect cost. A generated diff still needs review and project checks.
Selected source read 2026-10-20Capability sources read 9 October 2026. These are documentation observations; no product test was performed.
Research scope & testing evidence
Documentation update · 28 October 2026
Code-defined agents have distinct deployment and ownership roles. The current SDK package instructions differ from the launch post. Read the scoped update and sources.
This announcement check does not renew the plan prices or product-testing record.
Cloud canvases · 17 October source note
Cursor’s 16 October update adds cloud-agent canvases and a beta Environments API. Shared canvases require eligible paid access, team membership and compatible privacy settings. Environments support up to 300 repositories, subject to source-control access. Sharing and environment limits. No new price, speed test or comparative verdict.
How we would evaluate it
Cursor fits a developer who wants the agent close to the files, diffs and terminal they already inspect. Its value is shorter loops between a proposed change and your own verification. That also creates a temptation to accept a large diff because the agent produced it quickly. Start with a bounded fix, require the existing tests and keep architecture decisions in repository instructions. Budget model usage against real tasks, especially when moving from small local edits to longer cloud work; the subscription headline alone does not describe the whole allowance.
Editorial judgement. This analysis draws from the linked documentation and the workflow described here. It is not a benchmark or a report of using the product.
Fit and limits
Fit and limits
Check before committing
Model allowances and optional on-demand usage affect cost. A generated diff still needs review and project checks.
What the documentation confirms
What the documentation confirms
- The documentation covers local editor work, CLI, cloud agents, rules and review.
- Multiple model choices and usage pools are documented separately from subscriptions.
Source read 9 October 2026. Feature access can depend on plan, region and account.
Price and billing scope
Official pricing ↗Source read · USD
Public monthly Individual/Teams pricing; page offer metadata identifies USD. Annual/regional quotes not checked.
- FreeFree
Limited agent and chat requests.
- PlusUSD 8/month
Monthly billing; shared pool, Grok models only.
- PremiumUSD 30/month
Monthly billing; third-party models billed separately at API list price.
- SuperUSD 100/month
Monthly billing; larger shared usage pool.
- UltraUSD 200/month
Monthly billing; larger shared usage pool.
- TeamsFrom USD 30/user/month
Starting price, monthly billing; seat tier and additional model usage affect the bill.
- EnterpriseContact sales
Custom contract.
Taxes, fees and third-party model use are separate. Source does not establish when the old Pro/Standard plans changed. Source receipt.
Recorded editorial criteria
Recorded editorial criteria
These are the previously recorded editorial opinions used in the directory. They have no individually recorded assessment date and were not rescored during this documentation refresh. They are not product-test results.
| Functionality | 9/10 |
|---|---|
| Pricing value | 7/10 |
| Ease of use | 7/10 |
| Reliability | 8/10 |
| Founder fit | 9/10 |
Provider update · 14 October 2026
Cursor opened Origin App Marketplace submissions to the public, with approval required before listing. Read the announcement scope.
Edited by Clinton Feyisitan · Review scope: official announcement dated 2026-10-13, read 14 October 2026. No account or product workflow was tested. Existing price observations and editorial judgments keep their original dates.
Try one complete task
Use an existing failing test. Ask for the smallest fix, inspect every changed file and run the same tests before and after.
Create a trial worksheet →Compare requirements
Edited by Clinton Feyisitan · Evidence policy · Suggest a correction