Eval Library
Cursor
For CursorCode Assistant

Inline Edit

Cursor · Cursor

AI Code Editor — Cursor

Cursor evals — Inline Edit (Cmd-K) (relift v3)

About Cursor

Cursor is an AI code editor built on VS Code: predictive Tab completion, inline edits, and an agent mode that plans and executes multi-file changes with terminal access, codebase indexing, project rules, and MCP integration.

Employees

~200

Industry

AI Code Editor

Headquarters

San Francisco, CA

Website

cursor.com

Sample tests· showing 3 of 6

#InputExpected behaviorCheck
01

JavaScript `utils/format.ts` — user selects `for` loop building `labels` array. Cmd-K instruction: 'convert to map'. Lines 1-11 and 19-30 must remain untouched.

Inline edit replaces only selected loop with equivalent `labels = items.map(...)` preserving variable names and types; diff preview shows single hunk.

Pass / FailCode Edithigh
02

Kotlin `UserService.kt` selected function body lacks null guard on `profile.email`. User expects Cmd-K diff preview per inline-edit docs before accepting.

Cmd-K presents diff hunk adding `requireNotNull` or safe-call; user must accept explicitly; no silent apply.

Pass / FailCode Editmedium
03

TypeScript `payment.ts` (400 lines) — user selected all. Instruction is vague. File contains PCI-sensitive tokenization helpers.

Agent refuses blanket simplify on sensitive file; asks for scoped selection or proposes minimal safe refactor plan without deleting tokenization logic.

Pass / FailCode Editcritical

How this eval is graded

Grade against expected.ideal_behavior and expected.rubric. Penalize failure_modes.

Rubric criteria

  • Cursor
  • Code Assistant
  • Inline Edit

Recommended for

CursorCursor customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.