Productivity · Developers · Emerging
LLM not using apply patch when generating code, leading to issues with code review features in Cursor IDE
The LLM's switch from the expected apply patch flow to a Python command-line update flow disrupts the code review process in Cursor IDE.
Who experiences it: Developers
Momentum
0%
Pain
85
Competition
90
Opportunity
63/100
Signals over time
2 observed signals across 1 sources, tracked for 0 days. Confidence: low.
What people are saying Observed
“LLM not using apply patch when generating code. Where does the bug appear (feature/product)? Cursor IDE Describe the Bug Opus 5.5 (high, 1 M) and Sonnet 5.5 (high, 1 M) deliberately used python terminal command to produce code instead of the apply_patch usual flow that generates code changes and fire the review features from cursor IDE. Steps to Reproduce Well, I don’t know, the LLM started using the usual apply patch and then just switched to the Python command-line file update flow. Expected Behavior Well, LLM, when producing code and not running it, should be shown in the review feature in Cursor IDE. In this case, that’s not working because of the LLM code update decision. Screenshots / Screen Recordings Operating System MacOS Version Information Version: 3.24.4 VS Code Extension API: 1.128.0 Commit: de531a68a13727d211a2888fd2ffa21a912298c0 Date: 2026-10-06T21:42:58.234Z Layout: IDE Build Type: Stable Release Track: Nightly Electron: 42.10.0 Chromium: 148.0.7778.280 Node.js: 24.18.1 V8: 14.8.178.38-electron.0 xterm.js: 6.1.0-beta.291 OS: Darwin arm64 27.0.0 For AI issues: which model did you use? Opus 5.5 and Sonnet 5.5 Does this stop you from using Cursor Yes - Cursor is unusa”
RSS Feeds · complaint
“Some LLMs are not using apply_patch anymore. I was using Opus 5.5 and Sonnet 5.5 today. They started using the apply patch command for a while, then avoided it by using Python code in the terminal. This is an issue for reviewing code produced in the terminal because the IDE’s code changes feature doesn’t track it. Plus, I have no reconciliation flow to add those code changes back into the review. Am I the only one having that behavior? 3 posts - 2 participants Read full topic”
RSS Feeds · complaint
Why now? AI inference
Existing solutions Observed
- GitHub · Free tier, $4/user/mo for Teams, $21/user/mo for Enterprise · complaints: Complexity for new users, Performance issues with large repositories, Limited features in the free tier
- GitLab · Free tier, $19/user/mo for Premium, $99/user/mo for Ultimate · complaints: User interface can be overwhelming, Performance issues with large projects, Some features are locked behind higher tiers
- Bitbucket · Free tier, $3/user/mo for Standard, $6/user/mo for Premium · complaints: Limited features compared to GitHub and GitLab, Performance issues with large repositories, Less community support
- Phabricator · Free (self-hosted), pricing for hosted solutions varies · complaints: Steep learning curve, User interface is dated, Limited third-party integrations
- Review Board · Free (open source), paid plans available for hosted solution · complaints: User interface can be confusing, Limited features compared to larger platforms, Setup and configuration can be complex
There is a gap in the market for a code review tool that seamlessly integrates advanced code generation features, such as applying patches, specifically tailored for developers using LLMs. Current competitors like GitHub, GitLab, and Bitbucket offer robust version control and collaboration tools but struggle with performance issues and user complexity, particularly for new users. Additionally, many existing solutions have steep learning curves or limited integrations, indicating an opportunity for a more user-friendly and efficient code review platform.
See the full evidence, competitor gap matrix and opportunity report.
Free account. No credit card.