diff --git a/docs/CAVEATS.md b/docs/CAVEATS.md
index fdfbfce..bc697ca 100644
--- a/docs/CAVEATS.md
+++ b/docs/CAVEATS.md
@@ -38,6 +38,14 @@ This is a deliberate semantic misuse of the protocol. The debt it incurs:
 
 cosmic-comp makes the exact same bet (`clip_floating_windows` default-on in `AppearanceConfig`, `src/shell/element/window.rs:204`) and has carried the same complexity for years. This is a settled hack in Wayland-land, not a novel misstep — but it's still a hack. If a future protocol extension exposes "suppress client chrome" as a first-class signal, migrate to it and delete all of the above.
 
+## xcursor `pixels_rgba` is actually BGRA
+
+The `xcursor` crate's `Image::pixels_rgba` field is misleadingly named. The bytes come straight from the XCursor file, which stores pixels as `uint32` ARGB little-endian — i.e. `[B, G, R, A]` in memory. Interpreted as RGBA, the channels are wrong.
+
+The matching DRM fourcc for that byte order is `Fourcc::Argb8888` (which smithay maps to GL `BGRA_EXT`), **not** `Fourcc::Abgr8888`. Using `Abgr8888` swaps R and B on screen — a yellow cursor renders mint-blue, a red cursor renders violet, etc.
+
+niri gets this right in `src/cursor.rs` (`MemoryRenderBuffer::from_slice(&frame.pixels_rgba, Fourcc::Argb8888, ...)`). Do the same here.
+
 ## What to unit test
 
 Smithay glue code (handlers, delegates) is not worth testing — it's framework boilerplate. Write tests for **your** logic:
diff --git a/src/state/cursor.rs b/src/state/cursor.rs
index 7896c34..b6e21ac 100644
--- a/src/state/cursor.rs
+++ b/src/state/cursor.rs
@@ -82,7 +82,7 @@ impl CursorState {
                 }
                 let buffer = MemoryRenderBuffer::from_slice(
                     &img.pixels_rgba,
-                    Fourcc::Abgr8888,
+                    Fourcc::Argb8888,
                     (img.width as i32, img.height as i32),
                     1,
                     Transform::Normal,