Cursor 2.7.0's Glass layout replaces the secondary sidebar, forcing Codex and Claude extensions to the left Activity Bar. Here's what changed and the workaround.
Cursor's flat $0.25/M token surcharge on Luna marks up cache reads 1250% over OpenAI's base rate. Enterprise users with heavy context reuse bear the brunt.
A developer running Cursor, Codex, and Claude Code on one repo finds the real cost isn't subscriptions — it's the manual context handoff between tools.
A senior dev spent 20 minutes reviewing Claude Code's cleaner utility vs 5 minutes on a junior's messier PR. The bottleneck isn't quality — it's context loss.
An intern's Claude-generated code tripled a team's standup from 15 to 45 minutes. The real cost of AI coding slop is the review tax on senior developers.
Cursor's Bugbot code review agent drops from 5 minutes to 90 seconds per review, cuts costs 22%, and adds a /review command for pre-push checks.
Cursor 1.4 lets you steer the agent mid-task without waiting for it to finish, adds usage tracking in settings, and ships improved tool reliability.
Cursor 1.6 adds slash commands stored as Markdown files, conversation summarization, and terminal improvements for Agent mode.
Cursor 2.1 ships interactive plan clarifications, AI-powered code review inside the editor, and instant grep. Here's what changed.
Cursor 3.5 moves Automations into the Agents Window and adds multi-repo and no-repo configurations. New automations get 50% off agent runs for 7 days.
Claude Opus 4.8 is live in Windsurf with unchanged standard pricing at $5/$25 per million tokens and a significantly cheaper Fast Mode option.
Demo rewrite of the Cursor 2.5 article in the new news_brief style — one take, no H2s, table + warning callout.
Demo of the new news_brief format — short, opinionated, with markdown visual layers. Not a real release.
Cursor lets Teams admins and Individual plan users configure how thoroughly Bugbot reviews pull requests, with three effort levels now available.
Cursor added over 30 plugins from partners like Datadog, GitLab, and Hugging Face, letting the AI IDE read, write, and act across more of your stack.
Cursor's Bugbot can now learn rules in real time, connect to MCP servers, and ships its highest bug resolution rate yet.
Cursor's Composer 2 is now available, promising stronger results on challenging coding tasks. Here's what changed and who should care.
Cursor 3.2 adds /multitask for parallel async subagents, improved worktrees, and multi-root workspace support for cross-repo changes.
Cursor's latest release adds native PR review, parallel agent execution for faster plan runs, and quick-action pills for common workflows.
Cursor added a context usage breakdown for agent sessions. Here's what it shows and why it matters for debugging long AI coding sessions.
Windsurf launched Adaptive, an intelligent model router that picks the right AI model per task to reduce quota burn. Max users also lose daily limits.
Windsurf's updated model picker organizes models by family, adds hovercards for variant toggles like reasoning effort and speed, and lets you pin favorites.
Windsurf's latest update adds quota billing support and shows daily/weekly AI usage directly in the IDE. Here's what changed and why users are upset.
Cursor 2.5 adds a plugin system, async subagents that run tasks in parallel, and a sandbox.json networkPolicy that enables per-domain network access for sandboxed commands.
Cursor 2.6 adds interactive UI components inside agent chats via MCP Apps, plus team-scoped plugin marketplaces for sharing private tools across organizations.
Windsurf launched Adaptive, a model router that picks the right LLM per task. Plus: a redesigned model picker and no daily limits for Max users.
Windsurf 2.0 adds Claude Opus 4.7, Anthropic's most capable model for coding and agentic tasks, with improved sustained reasoning for long sessions.
Cursor 1.7 adds browser screenshots, Plan Mode, and Hooks — giving the AI agent eyes on your frontend and more structured coding workflows.
Cursor's latest release adds Debug Mode for reproducing tricky bugs, multi-agent judging for better code quality, improved Plan Mode, and pinned chats.
Cursor's CLI update adds foreground, background, and cloud-based agent modes to the terminal, closing the gap with CLI-native AI coding tools.
Cursor's cloud agents gained computer use — they can now autonomously run, test, and interact with the software they create, closing the build-test loop.
Cursor's latest release adds model controls, spend management, and detailed usage analytics for Enterprise admins managing AI coding costs.
Cursor launched Security Review in beta for Teams and Enterprise plans, adding two always-on AI agents that scan PRs for vulnerabilities and risky code patterns.
Cursor's team marketplace no longer requires connecting a repository first. Admins can manage first-party plugins directly from settings.
Cursor 2.0 introduces a proprietary coding model and a redesigned agent sidebar with plan-based workflows. Here's what shipped and how to use it.
Cursor now works inside JetBrains IDEs through the Agent Client Protocol. IntelliJ, PyCharm, and WebStorm users can access Cursor's AI features natively.
Cursor's new long-running agents can plan and execute complex coding tasks autonomously over extended periods, now available as a research preview.
Cursor shipped self-hosted cloud agents that keep code inside your infrastructure. Here's what changed, who it's for, and how to evaluate it.
Cursor announced an SDK for building programmatic coding agents. Here's what the changelog reveals, what's unconfirmed, and what to do next.
Windsurf lists a model called GPT-5.4 Mini at a 1x promotional credit rate. No OpenAI model matches that name. Here's what we know and don't.
Windsurf's AI-native IDE now shows a "GPT-5.5" option in its model dropdown. What we know, what we don't, and how to evaluate it.
Windsurf's changelog adds a model labeled GPT-5.4 with four reasoning tiers at 1x–3x credit costs. Here's what the listing says and how to use it.
A practical comparison of Cursor and Windsurf — two leading AI-powered code editors. Feature differences, pricing, and which one fits different developer profiles.