Cursor adds local Bugbot runs and selective PR change reviews
TL;DR
Bugbot now runs on Composer 2.5 for 3x faster performance at 22% lower cost. Users run Bugbot and security reviews locally via /review and limit to new changes since the last review.
What changed
Developers can now run Bugbot locally in Cursor with the /review command on Composer 2.5 to handle Security Reviews before pushing. Vibe Builders and Basic Users gain options to configure reviews that focus only on new PR changes since the last pass. The update runs three times faster at twenty two percent lower cost.
Why it matters
Developers see quicker iteration on code reviews with the speed and cost gains from Composer 2.5 compared to prior setups. Basic Users complete focused PR checks in less time during daily workflows while Vibe Builders maintain momentum across multiple iterations.
What to watch for
Track performance against alternatives like standard GitHub PR reviews on sample repositories. Verify by executing the /review command locally on a test branch and confirming it surfaces only the delta changes.
Who this matters for
- Vibe Builders: Use the /review command to check only new changes and keep your project momentum high.
- Developers: Run Bugbot locally with Composer 2.5 to catch security issues before you push to your repository.
Harsh’s take
Cursor is tightening the loop between writing code and shipping it. Moving Bugbot and Security Review locally via the /review command removes the friction of waiting for CI/CD pipelines to trigger remote checks. The 3x speed increase in Composer 2.5 makes local linting and security scanning feel like a real-time assistant rather than a heavy batch process.
The ability to review only the delta since the last pass is the real win for productivity. It prevents the cognitive fatigue of re-scanning old code and keeps the focus on the immediate task. This update signals a shift where the IDE is no longer just a text editor but a local quality assurance hub.
Operators should adopt this local-first workflow to reduce cloud compute costs and shorten their feedback cycles.
by Harsh Desai
About Cursor
View the full Cursor page →All Cursor updatesGo deeper
More AI news
- Daily RoundupQwen 3.8 debut, lipsync model on Replicate, and agent tools for daily builds
Vendors released new models and infrastructure updates while builders gained fresh agent and coding tools across Replicate, Hugging Face, and Product Hunt on 3 August.
- Weekly DigestHermes Agent v0.19.1 with GPT-5.6 and 80% latency cut, OpenClaw MCP apps (agent routing and recovery tools)
Hermes Agent rolled out v0.19.1 with new model support, durable ledgers, secret managers, and major speed gains while OpenClaw added MCP apps, structured questions, and local inference options across the same period.
- Daily RoundupCadence-fast on Replicate, Qwythos-27B on HF, and voice agents ready today
Fresh models appeared on Replicate and Hugging Face while Product Hunt surfaced local and voice agent tools, and Altman urged slower AI progress.