Cursor Review: What It Does, Pricing, and Alternatives

Draft v0.1 — 2026-05-23 KST. content_status = qa_passed. Generated from templates/tool-page-template.md. Promoted to qa_passed after a 2026-05-23 page-body read of cursor.com/pricing and cursor.com/; Section A of qa/adsense-seo-quality-gate.md satisfied. Meta description (≤ 155 chars): Cursor is Anysphere's AI-first code editor with agents, codebase chat, and Tab completion — here is what it does, what it costs, and how it compares.

Quick verdict

• Best for: developers who want an AI-first editor with agentic multi-file edits, codebase-wide chat, and a strong autocomplete experience as the default workflow — not as a sidecar extension on a general editor.
• Not ideal for: non-developers, teams already standardized on a different in-IDE AI assistant (notably GitHub Copilot inside an existing JetBrains/VS Code/Visual Studio workflow), and anyone whose top requirement is a chat assistant rather than an editor.
• Pricing model: freemium. Hobby is free, Individual is $20/month (with Pro, Pro+, and Ultra usage variants), Teams is $40/user/month, and Enterprise is Custom — verified on cursor.com/pricing on 2026-05-23.
• Free plan: yes — the Hobby plan is listed as Free with "No credit card required," "Limited Agent requests," and "Limited Tab completions." Specific numeric request and completion quotas were not exposed in the visible pricing page on 2026-05-23 and should be verified on the official site.
• Last verified: 2026-05-23 (cursor.com/pricing and cursor.com/ page-body reads)

Source-freshness note (2026-06-13)

If you are weighing Cursor as an AI coding-agent workflow rather than a generic editor add-on, a 2026-06-13 KST recheck confirms Cursor's official surfaces are live and positioned around an in-editor coding agent: the homepage https://www.cursor.com/ resolves to https://cursor.com/ (title "Cursor: AI coding agent"), and the official pricing page https://www.cursor.com/pricing resolves to https://cursor.com/pricing (title "Cursor · Pricing") and was reachable in the same pass.

• Workflow fit. Cursor presents itself as an AI coding-agent/editor surface — the agent works inside the editor where you already read and change code, not only as a separate chat window. Evaluate it against how your team actually reviews repositories and applies changes.
• Repository changes need human review. A coding agent proposes edits across your files; those changes still need a human to review the diff, test, and approve before they merge or ship. Treat agent output as a draft, not a finished commit, and decide up front where the agent runs and which parts of your codebase and tooling it is allowed to reach.
• Verify pricing and plans on the official site. The pricing page was reachable on 2026-06-13, but this note quotes no price, quota, plan entitlement, model availability, or feature limit — those shift between releases. Confirm current pricing, plan inclusions, and model availability on Cursor's official pricing page before relying on specifics.

Source-backed freshness note drawn from Cursor's own homepage and pricing page (reachability/title/positioning only). No benchmark, ranking, price, quota, speed, model-availability, or superiority claim is made here; vendor positioning is reported as vendor evidence only, not an independent ranking.

Source-freshness note (2026-06-18)

If your real question is whether Cursor fits an agent-driven coding workflow — not just whether it is a nice editor — a 2026-06-18 KST recheck of Cursor's own surfaces is a useful self-verification starting point. The official changelog (https://cursor.com/changelog, title "What's New in Cursor — Latest Updates and Release Notes") is live and currently leads with cloud-environment and cloud-subagent work in the Agents window, and its release notes describe agent, code-review, background, security, and enterprise surfaces alongside MCP. The official documentation (https://docs.cursor.com/, resolving to https://cursor.com/docs, title "Cursor Docs — Agent, Rules, MCP, Skills and CLI"), the homepage (https://cursor.com/, title "Cursor: AI coding agent", positioned as "your coding agent for building ambitious software"), and the pricing page (https://cursor.com/pricing, title "Cursor · Pricing") were all reachable in the same pass.

• Agent and cloud-subagent workflow. Cursor positions its agent work — including cloud environments and subagents running in the Agents window — as part of how you build, not as a side chat. Evaluate that against where your team actually wants an agent to run (locally vs. in a cloud environment) and how much of a task you are comfortable handing to a background agent before a human looks at the result.
• Review boundary stays with a human. A coding agent, cloud or local, proposes changes; those changes still need a person to review the diff, test, and approve before they merge or ship. Decide up front which parts of your codebase, sessions, and tooling the agent is allowed to reach, and treat the agent's code review surface as an aid to human review, not a replacement for it.
• Rules, MCP, CLI, and docs are where the setup lives. Cursor's own docs organize the configurable pieces — rules, MCP connections, skills, and the CLI — that determine how the agent behaves and what it can touch. Read those before granting broad scope, and confirm current pricing, plan inclusions, and model availability on Cursor's official pricing page rather than relying on any figure quoted here.

Source-backed freshness note: a 2026-06-18 KST reachability/title/positioning recheck of Cursor's own changelog, docs, homepage, and pricing pages. No price, quota, plan entitlement, model availability, benchmark, ranking, speed, or superiority claim is drawn from it, and no independent validation is implied — vendor positioning is reported as vendor evidence only.

Source-freshness note (2026-06-24)

If you are evaluating Cursor as an agent-driven coding-and-editor workflow — and want to verify its plans yourself — a 2026-06-24 KST source gate confirms Cursor's own surfaces for doing that evaluation are reachable and stable. The homepage (cursor.com, titled "Cursor: AI coding agent", H1 "Cursor is your coding agent for building ambitious software", HTTP 200), the pricing page (cursor.com/pricing, titled "Cursor · Pricing", H1 "Pricing", HTTP 200), the changelog (cursor.com/changelog, titled "What's New in Cursor — Latest Updates & Release Notes", HTTP 200), and the documentation (docs.cursor.com, resolving to https://cursor.com/docs, titled "Cursor Docs — Agent, Rules, MCP, Skills & CLI", HTTP 200) all loaded as full pages in the same pass.

• Evaluate the agent and editor workflow at the source. Cursor's homepage continues to position the product as a coding agent that works inside the editor where you read and change code. Read that, the changelog, and the docs on Cursor's own pages and decide whether the workflow fits how your team actually reviews repositories and applies changes — don't infer current capabilities from this page.
• Verify plans and pricing on the pricing page, not here. The pricing page remained reachable (HTTP 200). It — not this page — is where to confirm current plan names, prices, quotas, included models, and per-tier feature gating before you rely on any figure, including the ones quoted lower on this page once they are more than ~90 days old.
• Track changes in the docs and changelog, and keep a human in the loop. New agent, rules, MCP, skills, and CLI behavior ships and changes on Cursor's official changelog and docs; confirm current behavior and scope there before granting broad access. Treat any Cursor output — completions, chat, or an agent's proposed multi-file edits — as a draft your team reviews, tests, and approves before it merges or ships. The vendor's positioning is not a guarantee about any specific run.

Source-backed freshness note: a 2026-06-24 KST reachability/title/H1 recheck of Cursor's own homepage, pricing, changelog, and docs surfaces (from the source gate). No price, quota, per-plan feature, model-availability, benchmark, ranking, speed, accuracy, or superiority claim is made here; volatile specifics are routed to Cursor's official pages, and vendor positioning is reported as vendor evidence only.

What is Cursor?

Cursor is an AI-first code editor built by Anysphere. The product's positioning, as written on the official homepage at https://cursor.com/ on 2026-05-23, frames it as a coding agent first and an editor second — the tagline reads: "Built to make you extraordinarily productive, Cursor is the best coding agent." That framing matters because it tells you Cursor is not designed as a chat tool that happens to know about code; it is designed as a development environment in which agentic AI is the primary input method, alongside the keyboard.

Cursor exposes several distinct AI surfaces inside the editor and around it. The homepage on 2026-05-23 calls out, among others: an Agents surface for autonomous multi-step work (planning, building, testing, and "demoing" a feature), a Tab autocomplete model for predicting and inserting the next code change, a chat experience with codebase-wide context (semantic search and codebase indexing), a Code Review / BugBot surface for pull-request review, a CLI for invoking agents outside the editor window, and integrations into terminal, Slack, and GitHub. The homepage also references multiple underlying model providers (OpenAI, Anthropic, Gemini, xAI), which is consistent with Cursor's public positioning as a model-agnostic editor rather than a wrapper around one vendor's model line.

Cursor is widely understood in the developer community to be built on the open-source code from VS Code (Microsoft's editor, MIT-licensed), which is reflected in the editor's UX and extension model. The cursor.com/ homepage on 2026-05-23 did not itself say the words "VS Code"; readers who care about the editor lineage should verify the current status of that claim directly on the official site and in the project's documentation rather than from third-party copy. What is visible on the homepage is a macOS download button; Cursor's other platform downloads (Windows, Linux) are linked from the Download page rather than the homepage hero and should be checked there for current availability.

• Vendor: Anysphere
• Official homepage: https://cursor.com/
• Category: AI Coding Assistants

Main use cases

• Use case 1 — Agentic multi-file edits inside the editor: describe an outcome in chat or in an agent prompt, and Cursor's Agent surface plans the change, edits across files, and proposes the diff. This is the workflow Cursor's own homepage describes when it calls itself "the best coding agent," and it is the workflow most often cited by Cursor users as the reason for switching from a plain editor plus separate chat tab. Expect the agent to need direction and review on non-trivial changes — agentic edits are proposals, not finished work.
• Use case 2 — Codebase chat and search: ask questions about a repository ("where do we handle auth?", "what calls this function?", "what does this module assume about its inputs?") and have the answer drawn from indexed code rather than a model's training. The homepage references semantic search and complete codebase indexing as part of the codebase-understanding feature set on 2026-05-23.
• Use case 3 — Inline completion via Tab: Cursor's Tab model predicts the next edit, not just the next token, which on real code looks like multi-line completions, refactor-aware suggestions, and "accept this whole change" prompts. This is the surface that competes most directly with traditional in-IDE AI completion tools like GitHub Copilot's completion mode.
• Use case 4 — Reviewing pull requests with BugBot: the homepage on 2026-05-23 also mentions a Code Review / BugBot surface for PR review. The exact behavior, the tiers it is available on, and any integration setup should be verified on the official site before relying on it.

Pricing and plans

The values below were read directly from cursor.com/pricing on 2026-05-23 KST. Cursor's plan structure has changed multiple times across releases (notably around how request and completion quotas are labeled), so reconfirm with the official pricing page before quoting these numbers more than ~90 days from now.

• Hobby — Free. "No credit card required," "Limited Agent requests," and "Limited Tab completions." Specific numeric Agent-request and Tab-completion limits were not visible on the pricing card on 2026-05-23 and should be verified on the official site before being quoted in collateral.
• Individual — $20 / month. Bundles Pro, Pro+, and Ultra usage tiers under the same Individual plan label on the pricing page; "Everything in Hobby, plus: Extended limits on Agent, Access to frontier models, MCPs, skills, and hooks, Cloud agents, Bugbot on usage-based billing." The pricing page also shows a Monthly / Yearly toggle; the Yearly equivalent monthly price was not asserted in this fetch and should be verified on the official site.
• Teams — $40 / user / month. "Everything on Individual, plus: Cloud agents with shared team context, Team-wide rules, skills, and automations, Security review agent, SAML/OIDC SSO + enforced team-level privacy mode, Team plugin marketplace, Usage analytics, Centralized team billing."
• Enterprise — Custom pricing. "Everything in Teams, plus: Pooled usage, Invoice/PO billing, SCIM seat management, AI code tracking API and audit logs, Granular admin and model controls, Priority support and account management, Bugbot on custom plan." Contact Sales pricing — verify on the official site.

Source: live page-body read of https://cursor.com/pricing on 2026-05-23 KST. The Monthly/Yearly toggle, exact Hobby request/completion limits, exact Pro/Pro+/Ultra usage caps inside the Individual plan, region-specific pricing, and any active promotions were not in scope of this fetch. Verify on the official site before quoting.

When evaluating Cursor for a team or organization, also verify directly:

• Which underlying model providers are routed to in which plan, since model-provider mix affects both data-handling and per-request cost.
• Data-handling and code-snippet retention policy per plan, including whether team-level privacy mode is enforced by default on the tier you would buy.
• Platform downloads (macOS, Windows, Linux) and current platform parity for the feature you actually care about (agents, code review, CLI).
• The active definition of "Agent requests," "Tab completions," and any other quota unit — these labels have been refined between Cursor releases.
• Regional plan availability and the active currency on the pricing page in your locale.

Pros

• AI is the default workflow, not a side panel. The whole UX is built around prompting and accepting agent edits, which removes the context-switch tax of a separate chat tab plus extension.
• Model-agnostic positioning means an organization is not locked into one model provider — the homepage on 2026-05-23 names OpenAI, Anthropic, Gemini, and xAI as routed providers.
• Free tier (Hobby) is real (no credit card required), so individual developers can evaluate the editor's agent and Tab surfaces on real work before committing to a paid plan.
• Pricing structure maps cleanly onto procurement: individual paid tier at $20/month, a Teams plan with SSO and team-wide rules at $40/user/month, and an Enterprise tier with audit logs and SCIM seat management.

Cons and caveats

• AI-generated code can be subtly wrong: mishandled edge cases, off-by-one errors, missed null checks, insecure defaults, or hallucinated APIs. Treat all Cursor suggestions, including agent multi-file edits, as proposals that require code review, tests, and (for security-sensitive code) targeted analysis. Adopting Cursor does not eliminate the need for human review.
• Code-generation tools have outstanding legal questions around training-data sourcing, license inheritance, and code attribution. This page does not assert any legal conclusion; consult counsel before relying on AI-generated code for license-sensitive work.
• Data-handling differs by plan tier and by routed model provider. The Cursor pricing page on 2026-05-23 references "enforced team-level privacy mode" only on the Teams tier and above. Read Cursor's official documentation on data handling for the specific plan you intend to buy.
• Cursor is a dedicated editor, not an extension. Adoption means switching the editor itself, which is a heavier change than installing a plugin in an editor a team already uses. Some teams will not want to make that switch even if Cursor's AI surfaces are stronger.
• Quotas are labeled qualitatively on the Hobby plan ("Limited Agent requests," "Limited Tab completions") rather than numerically on the public pricing page. Heavy free-tier users should verify limits directly on the official site.
• Outputs are not professional advice. Do not treat Cursor's agent or chat responses as a substitute for licensed legal, security, accounting, financial, or medical counsel.

Alternatives

• GitHub Copilot — better if your team already lives on GitHub and wants AI to appear inside an existing IDE (VS Code, JetBrains, Visual Studio, Neovim, Xcode, and others listed on the official plans page) rather than in a new editor. Copilot's tightest integration is with the GitHub repo, PR, and review surfaces.
• Claude — better if your top need is a general-purpose assistant for long-context reasoning, drafting, and code discussions across many tasks, not a dedicated editor; verify on official site whether you also need a separate in-editor tool for inline completion.
• Tabnine — better if your organization requires self-hosted, private-model, or strict enterprise data-isolation deployment that a hosted editor like Cursor does not provide. Verify Tabnine's current SKU lineup on the official site.
• Replit AI — better for browser-based development, education, hobbyist projects, and quick prototypes where the entire development environment lives in the browser rather than on a developer's machine.

Where to compare Cursor next

If Cursor is on your shortlist, the next question is usually "against what, and for which job?" These side-by-side pages are organized by workflow fit, not by a winner — each one walks through where one tool's shape suits a particular task better than the other, so you can follow the path that matches your own work.

Same-category coding choices — these are direct workflow alternatives to Cursor, where the decision is really about which coding surface you want to live in day to day:

• Cursor vs GitHub Copilot — when you are choosing between a dedicated AI-first editor and AI that appears inside an IDE you already use.
• Tabnine vs Cursor — when self-hosted, private-model, or strict data-isolation deployment is the deciding factor versus a hosted editor.
• Cursor vs Replit AI — when the question is a desktop editor on your machine versus a browser-based development environment.
• Cursor vs Claude — when you are weighing a dedicated coding editor against a general reasoning-and-coding chat assistant for your coding workflow.

Cross-category comparisons — these tools sit in writing, productivity, or automation rather than coding, so they are not direct substitutes for an editor; the pages are useful when your work spans more than one category:

• Cursor vs Jasper — coding editor versus a marketing-and-writing tool; different jobs, listed for teams comparing across categories.
• Cursor vs Notion AI — coding editor versus an in-document productivity assistant, not a like-for-like swap.
• Cursor vs Microsoft Copilot — coding editor versus an assistant embedded across Microsoft 365 productivity apps.
• Grammarly AI vs Cursor — a writing-assistance tool versus a code editor; relevant only when a single team owns both writing and coding.
• Zapier AI vs Cursor — app-automation versus an in-editor coding workflow; adjacent jobs rather than competing ones.

To browse the whole field rather than a single head-to-head, start from the AI Coding category. These links are decision paths, not rankings — no benchmark, price, quota, speed, or model-availability claim is made here; the comparison pages route any such specifics to the official sources.

Buyer control and the review boundary

<!-- marker: cursor-agent-review-boundary-2026-06-27 -->

If you are evaluating Cursor as a buyer rather than a solo user, the deciding question is usually not "can the agent write code?" but "who stays in control of the changes it writes, and where does the review boundary sit?" This page makes no benchmark, ranking, or superiority claim; it only frames the control questions to ask, and routes the answers to your own repository and review practice plus Cursor's official surfaces. <span id="cursor-agent-review-boundary-2026-06-27"></span>

• Who reviews AI-generated changes. Cursor's agent, completions, and chat all produce drafts, not finished commits. Decide in advance who reviews each diff — the prompting developer, a second reviewer, or both — and treat that human review as a required step rather than an optional one. The same caveats in Cons and caveats (subtly wrong edits, missed edge cases, insecure defaults) are why the reviewer, not the agent, owns whether a change is correct.
• Repository, PR, and test boundaries. Before granting scope, decide which parts of the codebase the agent may read and change, whether its output flows through your normal pull-request and code-review process, and which tests must pass before a change can merge. Keeping agent output inside the same PR and test gates you already use for human work is the simplest way to hold the review boundary in one place. Cursor's own code-review surface is an aid to that process, not a replacement for the human gate.
• When a coding-agent workflow needs human approval. A background or cloud agent that runs further from a developer's eyes raises the stakes of the same question: how much of a task are you comfortable handing off before a person looks at the result, and where is the explicit approve-before-merge step? For anything touching security-sensitive code, shared infrastructure, or release-bound branches, set human approval as a hard gate rather than a convention.
• How to compare against adjacent routes. The control/review boundary is a fair axis to compare tools on, not just a Cursor concern. Use the same questions when you weigh Cursor against an in-IDE assistant on Cursor vs GitHub Copilot, a self-hosted or data-isolation deployment on Tabnine vs Cursor, a browser-based environment on Cursor vs Replit AI, or a general reasoning-and-coding chat on Cursor vs Claude. To browse the field by job rather than a single head-to-head, start from the AI Coding category.

Evergreen decision framing only. No price, quota, plan entitlement, model availability, benchmark, ranking, speed, accuracy, security-certification, or legal claim is made here; verify current plan inclusions and data-handling on Cursor's official pages, and confirm how the agent fits your repository and review process against your own practice.

Who should not use Cursor

• Teams whose code license, compliance posture, or contractual obligations are incompatible with sending source code to a third-party AI editor. Verify Cursor's data-handling for the specific plan you would buy before adopting.
• Beginners who have not yet learned the underlying language or framework. Uncritical accept-all use of agent edits can cement subtle bugs and bad patterns, and reviewers downstream will not always catch them.
• Organizations that have already standardized on GitHub Copilot or another in-IDE AI assistant inside an existing editor and would only fragment workflows by adding Cursor in parallel.
• Developers whose primary work is in an editor Cursor does not target as a first-class platform; verify the Download page for current Mac/Windows/Linux parity before adopting.

Author selection rubric

Choose Cursor when at least two of these are true:

• You want AI to be the default input method in your editor, not a sidecar.
• You routinely make multi-file edits and want an agent that can plan and apply them across the codebase.
• You want a model-agnostic editor so you can route to OpenAI, Anthropic, Gemini, or xAI without re-tooling.

Avoid Cursor when any of these are true:

• Your team is committed to staying inside an existing editor (most often VS Code with extensions, JetBrains, or Visual Studio) and will not switch editors regardless of AI quality.
• Your top requirement is a general-purpose chat assistant rather than an editor; pay for a chat product like Claude or ChatGPT instead.
• Your data-handling, compliance, or licensing posture cannot be met by Cursor's current plan tiers — verify on the official site before adopting.

Sources

• Official homepage: https://cursor.com/ — recorded as src-cursor-needs-verify in data/sources.json with access_status = ok; cited only as the official product URL and for the homepage tagline / feature names verified on 2026-05-23.
• Official pricing page: https://cursor.com/pricing — recorded as src-cursor-pricing-2026-05-23 in data/sources.json with access_status = ok after a 2026-05-23 page-body read; this is the source for every plan name, price, and Free-tier feature wording quoted on this page.
• Official homepage (freshness recheck): https://cursor.com/ (reached from https://www.cursor.com/, title "Cursor: AI coding agent") — recorded as src-cursor-homepage-2026-06-13 in data/sources.json with access_status = ok after a 2026-06-13 KST HTTP 200 reachability/title/positioning read; source of the "Source-freshness note (2026-06-13)" section. The companion pricing page https://cursor.com/pricing (title "Cursor · Pricing") was confirmed reachable in the same pass; used as vendor evidence only, with no price, quota, plan, model-availability, benchmark, or ranking claim drawn from it.
• Official changelog and docs (freshness recheck): https://cursor.com/changelog (title "What's New in Cursor — Latest Updates and Release Notes") and https://docs.cursor.com/ → https://cursor.com/docs (title "Cursor Docs — Agent, Rules, MCP, Skills and CLI") — recorded as src-cursor-changelog-docs-2026-06-18 in data/sources.json with access_status = ok after a 2026-06-18 KST HTTP 200 reachability/title/positioning read; source of the "Source-freshness note (2026-06-18)" section. The homepage https://cursor.com/ (title "Cursor: AI coding agent") and pricing page https://cursor.com/pricing (title "Cursor · Pricing") were confirmed reachable in the same pass. Used as vendor evidence only — agent/cloud-subagent workflow, review boundary, security/session/tool scope, and rules/MCP/CLI/docs positioning — with no price, quota, plan, model-availability, benchmark, ranking, speed, or superiority claim drawn from it and no independent validation implied.
• Official source-gate reachability recheck (2026-06-24 KST): https://cursor.com/ (title "Cursor: AI coding agent", H1 "Cursor is your coding agent for building ambitious software"), https://cursor.com/pricing (title "Cursor · Pricing", H1 "Pricing"), https://cursor.com/changelog (title "What's New in Cursor — Latest Updates & Release Notes"), and https://docs.cursor.com/ → https://cursor.com/docs (title "Cursor Docs — Agent, Rules, MCP, Skills & CLI") — recorded as src-cursor-source-gate-2026-06-24 in data/sources.json with access_status = ok after 2026-06-24 KST HTTP 200 reads in the source gate (data/cursor-source-gate-2026-06-24-operator.json, records cursor_home, cursor_pricing, cursor_changelog, cursor_docs); source of the "Source-freshness note (2026-06-24)" section. Durable reachability/title/H1 evidence only — no price, quota, plan, model-availability, benchmark, ranking, speed, accuracy, or superiority claim is drawn from it; the 2026-05-23 page-body reads remain the source of every plan/price quoted on this page.
• Vendor: Anysphere — https://cursor.com/

Sources marked needs_verification or blocked in data/sources.json must be re-fetched live before publish. Note the recheck date in the update log.

Internal links (at least 3)

• Category page: /ai-coding/
• Alternative tool: /tools/github-copilot/
• Alternative tool: /tools/claude/
• Comparison pages: /compare/cursor-vs-github-copilot/, /compare/tabnine-vs-cursor/, /compare/cursor-vs-replit-ai/, /compare/cursor-vs-claude/, /compare/cursor-vs-jasper/, /compare/cursor-vs-notion-ai/, /compare/cursor-vs-microsoft-copilot/, /compare/grammarly-ai-vs-cursor/, /compare/zapier-ai-vs-cursor/

Disclosure

• Affiliate links: none.
• Sponsored content: none. Anysphere has no relationship to this page.
• Generative AI assistance: this draft was assembled with the help of an AI assistant working from a 2026-05-23 live read of the official Cursor homepage and pricing page; every plan, price, and feature claim is constrained to wording visible on those pages on that date.

Trademark notice

Cursor and Anysphere are trademarks of Anysphere. VS Code is a trademark of Microsoft. Use here is referential only and does not imply endorsement, partnership, or affiliation.

Update log

• 2026-05-23 (draft and qa pass): first local draft created from templates/tool-page-template.md. Live page-body reads of https://cursor.com/ and https://cursor.com/pricing on 2026-05-23 KST provided the Hobby / Individual ($20 / mo.) / Teams ($40 / user / mo.) / Enterprise (Custom) plan names and the Free-tier wording. New source entry added (src-cursor-pricing-2026-05-23, access_status = ok). data/tools.json pricing_summary, pricing_model, has_free_plan, platforms, confidence_score, last_verified_at, and content_status refreshed. data/sources.json updated. Section A1/A2 of qa/adsense-seo-quality-gate.md now satisfied; A3 (≥ 700 words), A4 (trust/safety/trademark), A5 (disclosure), and A6 (≥ 3 internal links) all pass. content_status advanced to qa_passed.
• 2026-06-13 (source-freshness refresh — local source, NOT a new revenue page): added a compact "Source-freshness note (2026-06-13)" section after "## Quick verdict" and before "## What is Cursor?", framing Cursor as an AI coding-agent/editor workflow — workflow fit (the agent runs inside the editor where code is read and changed), that repository changes still need a human to review the diff, test, and approve before they merge or ship (and that teams should decide where the agent runs and what it can reach), and that pricing/plan/model availability must be confirmed on Cursor's official pricing page. Evidence: a 2026-06-13 KST HTTP 200 recheck of https://www.cursor.com/ (final URL https://cursor.com/, title "Cursor: AI coding agent"; durable markers Cursor, AI, coding agent, agent, enterprise) plus the companion https://www.cursor.com/pricing (final URL https://cursor.com/pricing, title "Cursor · Pricing") confirmed reachable in the same pass. Recorded as one new source src-cursor-homepage-2026-06-13 (access_status = ok) added to data/sources.json and the cursor sources_used; a Sources bullet was added here. The page is treated as vendor evidence only, not an independent ranking. No price, quota, plan entitlement, model-availability, benchmark, ranking, speed, accuracy, or superiority claim was added (no price/quota wording is quoted); the prior pricing rows are unchanged; no Gumroad/affiliate/sponsored/coupon/UTM link was added; data/tools.json and last_verified_at are unchanged. content_type/content_status stay tool_page/qa_passed; revenue inventory unchanged (18 tool + 51 comparison = 69 qa_passed pages).
• 2026-06-18 (source-freshness refresh — local source, NOT a new revenue page, NOT a deploy): added a compact "Source-freshness note (2026-06-18)" section after the "Source-freshness note (2026-06-13)" section and before "## What is Cursor?", framing Cursor as an agent-driven coding workflow — agent/cloud-subagent workflow (cloud environments and subagents in the Agents window), the human-owned review boundary (review the diff/test/approve before merge or ship; decide which codebase/sessions/tooling the agent may reach; the code-review surface aids human review rather than replacing it), and that rules/MCP/CLI/docs are where the agent's behavior and scope are configured — with pricing/plan/model availability routed to Cursor's official pricing page. Evidence: a 2026-06-18 KST HTTP 200 reachability/title/positioning recheck of Cursor's own changelog https://cursor.com/changelog (title "What's New in Cursor — Latest Updates and Release Notes"; H1 on cloud environment setup and cloud subagents in the Agents window; agent/review/security/enterprise/background/MCP markers) and docs https://docs.cursor.com/ → https://cursor.com/docs (title "Cursor Docs — Agent, Rules, MCP, Skills and CLI"), with the homepage https://cursor.com/ (title "Cursor: AI coding agent") and pricing page https://cursor.com/pricing (title "Cursor · Pricing") confirmed reachable in the same pass. Recorded as one new source src-cursor-changelog-docs-2026-06-18 (source_type = changelog, access_status = ok) added to data/sources.json and the cursor sources_used; a Sources bullet was added here. Vendor evidence only, not an independent ranking and no independent validation implied. No price, quota, plan entitlement, model-availability, benchmark, ranking, speed, accuracy, or superiority claim was added (no price/quota wording is quoted); the prior pricing rows are unchanged; no Gumroad/affiliate/sponsored/coupon/UTM link was added; data/tools.json and last_verified_at are unchanged. content_type/content_status stay tool_page/qa_passed; revenue inventory unchanged (18 tool + 51 comparison = 69 qa_passed pages).
• 2026-06-24 (source-freshness/traffic refresh — local source-gate recheck, NOT a new revenue page, NOT a deploy): added a compact "Source-freshness note (2026-06-24)" section after the "Source-freshness note (2026-06-18)" section (before "## What is Cursor?"), framing Cursor as an agent-driven coding-and-editor workflow whose plans should be verified at Cursor's own pages, with agent/editor output still needing human review and testing and no independent ranking or benchmark. Evidence: the 2026-06-24 KST source gate (data/cursor-source-gate-2026-06-24-operator.json) found Cursor's official surfaces reachable (HTTP 200) — https://cursor.com/ (title "Cursor: AI coding agent", H1 "Cursor is your coding agent for building ambitious software", body_sha256 prefix 1b1705cf06abf0b3), https://cursor.com/pricing (title "Cursor · Pricing", H1 "Pricing", prefix 50a1676a1047bb04), https://cursor.com/changelog (title "What's New in Cursor — Latest Updates & Release Notes", prefix f4968c9b4e42c16e), and https://docs.cursor.com/ → https://cursor.com/docs (title "Cursor Docs — Agent, Rules, MCP, Skills & CLI", prefix dd40a7c972bd44ee). Recorded as src-cursor-source-gate-2026-06-24 (source_type = official_homepage, access_status = ok) added to data/sources.json and the cursor sources_used, plus a Sources bullet. Durable reachability/title/H1 evidence only: no exact price, quota, per-plan feature, model-availability, benchmark, ranking, speed, accuracy, or superiority claim was added, and the 2026-05-23 page-body reads remain the source of every plan/price quoted here. No Gumroad/UTM/affiliate/sponsored/coupon link added; data/tools.json and last_verified_at unchanged; content_type/content_status stay tool_page/qa_passed; revenue inventory unchanged (18 tool + 51 comparison = 69 qa_passed pages). Local increment only — not a deploy.
• 2026-06-27 (qualified-traffic/decision-content increment — evergreen, NOT a new revenue page, NOT a deploy): added a compact "Buyer control and the review boundary" section after "## Where to compare Cursor next" and before "## Who should not use Cursor", carrying the durable marker cursor-agent-review-boundary-2026-06-27 (HTML comment + <span id> anchor). The section helps AI Stack DB readers evaluate Cursor by control/review boundaries — who reviews AI-generated changes, how repository/PR/test boundaries are handled, when a coding-agent workflow needs human approval, and how to apply the same control axis when comparing against adjacent existing routes. Links only existing local routes already in this content package: /compare/cursor-vs-github-copilot/, /compare/tabnine-vs-cursor/, /compare/cursor-vs-replit-ai/, /compare/cursor-vs-claude/, and the /ai-coding/ category, plus an in-page anchor to "## Cons and caveats". No source was fetched. No price, quota, plan entitlement, model availability, benchmark, ranking, speed, accuracy, superiority, security-certification, or legal claim was added; no Gumroad/UTM/affiliate/sponsored/coupon/checkout link added. data/tools.json, data/sources.json, last_verified_at, and content/content-status.json are unchanged; content_type/content_status stay tool_page/qa_passed. Revenue inventory unchanged (18 tool + 51 comparison = 69 qa_passed pages). Local increment only — not a deploy.
• 2026-06-04 (internal-link/traffic refresh — live deployed later, NOT a new revenue page): added a "Where to compare Cursor next" section (placed after "## Alternatives", before "## Who should not use Cursor") linking nine now-live qa_passed Cursor comparison pages — /compare/cursor-vs-github-copilot/, /compare/tabnine-vs-cursor/, /compare/cursor-vs-replit-ai/, /compare/cursor-vs-claude/, /compare/cursor-vs-jasper/, /compare/cursor-vs-notion-ai/, /compare/cursor-vs-microsoft-copilot/, /compare/grammarly-ai-vs-cursor/, /compare/zapier-ai-vs-cursor/ — plus the /ai-coding/ category hub. Same-category coding comparisons (GitHub Copilot, Tabnine, Replit AI, Claude) are framed as direct workflow choices; writing/productivity/automation comparisons (Jasper, Notion AI, Microsoft Copilot, Grammarly, Zapier) are framed as cross-category, not direct substitutes. Updated the "Internal links" comparison row to list these nine comparisons (it previously listed only /compare/cursor-vs-github-copilot/). No source was fetched and no volatile claim was added: no benchmark, ranking, price, quota, model-availability, speed, or accuracy fact is asserted, and no commercial, affiliate, sponsored, coupon, tracking-parameter, or sales CTA link was added. data/sources.json, data/tools.json, content-status.json, and last_verified_at are unchanged. Hermes later live-smoked /tools/cursor/ with the new section present; this 2026-06-11 hygiene pass removes stale pre-deploy wording from the generated page body.