v1.4.12
AIPentest/CyberStrikeAIv1.4.12Apr 9, 2026by Ed1s0nZ
AI Summary
Significant performance improvements were made to the chat page by eliminating N+1 API requests, batching data loading, optimizing message rendering, and improving backend search query efficiency.
Key Highlights
- Elimination of N+1 API request patterns across the chat page.
- Batch loading of group mappings and tool name resolution.
- Parallel execution of page initialization requests.
- Optimized backend search queries using EXISTS subqueries.
New Features
- Batch group mapping loading
- Batch tool name resolution
- Optimized message rendering
- Optimized backend search queries
Full Release Notes
## Performance Optimization Summary Eliminated N+1 API request patterns across the chat page. Group mappings now load in a single batch request instead of one per group. Tool name resolution for execution buttons is batched into one POST instead of individual GETs per tool. Page initialization parallelizes three independent API calls via `Promise.all`. Conversation switching no longer triggers a redundant full sidebar reload. Messages render in batches using `requestAnimationFrame` to avoid blocking the main thread. Backend search queries use `EXISTS` subqueries instead of `LEFT JOIN + DISTINCT`. ## 性能优化总结 消除了对话页面的 N+1 请求问题。分组映射改为单次批量接口获取；工具名称从逐个请求改为一次批量获取；页面初始化三个独立请求改为 `Promise.all` 并行；切换对话不再触发冗余的侧边栏全量刷新；消息渲染改为 `requestAnimationFrame` 分批执行，避免阻塞主线程；后端搜索从 `LEFT JOIN + DISTINCT` 改为 `EXISTS` 子查询。