Fix merge issues

Previously pressing tab would switch TUI focus to the history scrollbox - no longer necessary.

codex-rs/Cargo.lock

@@ -186,6 +186,26 @@ version = "1.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "dde20b3d026af13f561bdd0f15edf01fc734f0dafcedbaf42bba506a9517f223"
 
+[[package]]
+name = "arboard"
+version = "3.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "55f533f8e0af236ffe5eb979b99381df3258853f00ba2e44b6e1955292c75227"
+dependencies = [
+ "clipboard-win",
+ "image",
+ "log",
+ "objc2",
+ "objc2-app-kit",
+ "objc2-core-foundation",
+ "objc2-core-graphics",
+ "objc2-foundation",
+ "parking_lot",
+ "percent-encoding",
+ "windows-sys 0.59.0",
+ "x11rb",
+]
+
 [[package]]
 name = "arg_enum_proc_macro"
 version = "0.3.4"
@@ -926,6 +946,7 @@ name = "codex-tui"
 version = "0.0.0"
 dependencies = [
  "anyhow",
+ "arboard",
  "base64 0.22.1",
  "chrono",
  "clap",
@@ -1405,6 +1426,16 @@ dependencies = [
  "winapi",
 ]
 
+[[package]]
+name = "dispatch2"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "89a09f22a6c6069a18470eb92d2298acf25463f14256d24778e1230d789a2aec"
+dependencies = [
+ "bitflags 2.9.1",
+ "objc2",
+]
+
 [[package]]
 name = "display_container"
 version = "0.9.0"
@@ -1868,6 +1899,16 @@ dependencies = [
  "version_check",
 ]
 
+[[package]]
+name = "gethostname"
+version = "0.4.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0176e0459c2e4a1fe232f984bca6890e681076abb9934f6cea7c326f3fc47818"
+dependencies = [
+ "libc",
+ "windows-targets 0.48.5",
+]
+
 [[package]]
 name = "getopts"
 version = "0.2.23"
@@ -3064,6 +3105,42 @@ dependencies = [
  "objc2-encode",
 ]
 
+[[package]]
+name = "objc2-app-kit"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e6f29f568bec459b0ddff777cec4fe3fd8666d82d5a40ebd0ff7e66134f89bcc"
+dependencies = [
+ "bitflags 2.9.1",
+ "objc2",
+ "objc2-core-graphics",
+ "objc2-foundation",
+]
+
+[[package]]
+name = "objc2-core-foundation"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1c10c2894a6fed806ade6027bcd50662746363a9589d3ec9d9bef30a4e4bc166"
+dependencies = [
+ "bitflags 2.9.1",
+ "dispatch2",
+ "objc2",
+]
+
+[[package]]
+name = "objc2-core-graphics"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "989c6c68c13021b5c2d6b71456ebb0f9dc78d752e86a98da7c716f4f9470f5a4"
+dependencies = [
+ "bitflags 2.9.1",
+ "dispatch2",
+ "objc2",
+ "objc2-core-foundation",
+ "objc2-io-surface",
+]
+
 [[package]]
 name = "objc2-encode"
 version = "4.1.0"
@@ -3078,6 +3155,18 @@ checksum = "900831247d2fe1a09a683278e5384cfb8c80c79fe6b166f9d14bfdde0ea1b03c"
 dependencies = [
  "bitflags 2.9.1",
  "objc2",
+ "objc2-core-foundation",
+]
+
+[[package]]
+name = "objc2-io-surface"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7282e9ac92529fa3457ce90ebb15f4ecbc383e8338060960760fa2cf75420c3c"
+dependencies = [
+ "bitflags 2.9.1",
+ "objc2",
+ "objc2-core-foundation",
 ]
 
 [[package]]
@@ -5839,6 +5928,21 @@ dependencies = [
  "windows_x86_64_msvc 0.42.2",
 ]
 
+[[package]]
+name = "windows-targets"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9a2fa6e2155d7247be68c096456083145c183cbbbc2764150dda45a87197940c"
+dependencies = [
+ "windows_aarch64_gnullvm 0.48.5",
+ "windows_aarch64_msvc 0.48.5",
+ "windows_i686_gnu 0.48.5",
+ "windows_i686_msvc 0.48.5",
+ "windows_x86_64_gnu 0.48.5",
+ "windows_x86_64_gnullvm 0.48.5",
+ "windows_x86_64_msvc 0.48.5",
+]
+
 [[package]]
 name = "windows-targets"
 version = "0.52.6"
@@ -5877,6 +5981,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "597a5118570b68bc08d8d59125332c54f1ba9d9adeedeef5b99b02ba2b0698f8"
 
+[[package]]
+name = "windows_aarch64_gnullvm"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2b38e32f0abccf9987a4e3079dfb67dcd799fb61361e53e2882c3cbaf0d905d8"
+
 [[package]]
 name = "windows_aarch64_gnullvm"
 version = "0.52.6"
@@ -5895,6 +6005,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e08e8864a60f06ef0d0ff4ba04124db8b0fb3be5776a5cd47641e942e58c4d43"
 
+[[package]]
+name = "windows_aarch64_msvc"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dc35310971f3b2dbbf3f0690a219f40e2d9afcf64f9ab7cc1be722937c26b4bc"
+
 [[package]]
 name = "windows_aarch64_msvc"
 version = "0.52.6"
@@ -5913,6 +6029,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c61d927d8da41da96a81f029489353e68739737d3beca43145c8afec9a31a84f"
 
+[[package]]
+name = "windows_i686_gnu"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a75915e7def60c94dcef72200b9a8e58e5091744960da64ec734a6c6e9b3743e"
+
 [[package]]
 name = "windows_i686_gnu"
 version = "0.52.6"
@@ -5943,6 +6065,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "44d840b6ec649f480a41c8d80f9c65108b92d89345dd94027bfe06ac444d1060"
 
+[[package]]
+name = "windows_i686_msvc"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8f55c233f70c4b27f66c523580f78f1004e8b5a8b659e05a4eb49d4166cca406"
+
 [[package]]
 name = "windows_i686_msvc"
 version = "0.52.6"
@@ -5961,6 +6089,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8de912b8b8feb55c064867cf047dda097f92d51efad5b491dfb98f6bbb70cb36"
 
+[[package]]
+name = "windows_x86_64_gnu"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "53d40abd2583d23e4718fddf1ebec84dbff8381c07cae67ff7768bbf19c6718e"
+
 [[package]]
 name = "windows_x86_64_gnu"
 version = "0.52.6"
@@ -5979,6 +6113,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "26d41b46a36d453748aedef1486d5c7a85db22e56aff34643984ea85514e94a3"
 
+[[package]]
+name = "windows_x86_64_gnullvm"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0b7b52767868a23d5bab768e390dc5f5c55825b6d30b86c844ff2dc7414044cc"
+
 [[package]]
 name = "windows_x86_64_gnullvm"
 version = "0.52.6"
@@ -5997,6 +6137,12 @@ version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9aec5da331524158c6d1a4ac0ab1541149c0b9505fde06423b02f5ef0106b9f0"
 
+[[package]]
+name = "windows_x86_64_msvc"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ed94fce61571a4006852b7389a063ab983c02eb1bb37b47f8272ce92d06d9538"
+
 [[package]]
 name = "windows_x86_64_msvc"
 version = "0.52.6"
@@ -6057,6 +6203,23 @@ version = "0.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ea2f10b9bb0928dfb1b42b65e1f9e36f7f54dbdf08457afefb38afcdec4fa2bb"
 
+[[package]]
+name = "x11rb"
+version = "0.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5d91ffca73ee7f68ce055750bf9f6eca0780b8c85eff9bc046a3b0da41755e12"
+dependencies = [
+ "gethostname",
+ "rustix 0.38.44",
+ "x11rb-protocol",
+]
+
+[[package]]
+name = "x11rb-protocol"
+version = "0.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ec107c4503ea0b4a98ef47356329af139c0a4f7750e621cf2973cd3385ebcb3d"
+
 [[package]]
 name = "yaml-rust"
 version = "0.4.5"

codex-rs/tui/Cargo.toml

@@ -22,6 +22,7 @@ workspace = true
 
 [dependencies]
 anyhow = "1"
+arboard = "3"
 base64 = "0.22.1"
 chrono = { version = "0.4", features = ["serde"] }
 clap = { version = "4", features = ["derive"] }
@@ -40,7 +41,10 @@ codex-ollama = { path = "../ollama" }
 color-eyre = "0.6.3"
 crossterm = { version = "0.28.1", features = ["bracketed-paste"] }
 diffy = "0.4.2"
-image = { version = "^0.25.6", default-features = false, features = ["jpeg"] }
+image = { version = "^0.25.6", default-features = false, features = [
+    "jpeg",
+    "png",
+] }
 lazy_static = "1"
 once_cell = "1"
 mcp-types = { path = "../mcp-types" }

codex-rs/tui/src/app.rs

@@ -36,6 +36,55 @@ use std::time::Instant;
 /// Time window for debouncing redraw requests.
 const REDRAW_DEBOUNCE: Duration = Duration::from_millis(1);
 
+// Testable helper: generic over paste function so we can inject stubs in unit tests.
+fn try_handle_ctrl_v_with<F>(
+    app_event_tx: &AppEventSender,
+    key_event: &KeyEvent,
+    paste_fn: F,
+) -> bool
+where
+    F: Fn() -> Result<
+        (std::path::PathBuf, crate::clipboard_paste::PastedImageInfo),
+        crate::clipboard_paste::PasteImageError,
+    >,
+{
+    if key_event.kind == KeyEventKind::Press
+        && key_event.code == KeyCode::Char('v')
+        && key_event
+            .modifiers
+            .contains(crossterm::event::KeyModifiers::CONTROL)
+    {
+        match paste_fn() {
+            Ok((path, info)) => {
+                tracing::info!(
+                    "ctrl_v_image imported path={:?} width={} height={} format={}",
+                    path,
+                    info.width,
+                    info.height,
+                    info.encoded_format_label
+                );
+                app_event_tx.send(AppEvent::AttachImage {
+                    path,
+                    width: info.width,
+                    height: info.height,
+                    format_label: info.encoded_format_label,
+                });
+                return true; // consumed
+            }
+            Err(err) => {
+                tracing::debug!("Ctrl+V image import failed: {err}");
+            }
+        }
+    }
+    false
+}
+
+fn try_handle_ctrl_v(app_event_tx: &AppEventSender, key_event: &KeyEvent) -> bool {
+    try_handle_ctrl_v_with(app_event_tx, key_event, || {
+        crate::clipboard_paste::paste_image_to_temp_png()
+    })
+}
+
 /// Top-level application state: which full-screen view is currently active.
 #[allow(clippy::large_enum_variant)]
 enum AppState<'a> {
@@ -113,6 +162,9 @@ impl App<'_> {
                         if let Ok(event) = crossterm::event::read() {
                             match event {
                                 crossterm::event::Event::Key(key_event) => {
+                                    if try_handle_ctrl_v(&app_event_tx, &key_event) {
+                                        continue;
+                                    }
                                     app_event_tx.send(AppEvent::KeyEvent(key_event));
                                 }
                                 crossterm::event::Event::Resize(_, _) => {
@@ -499,6 +551,16 @@ impl App<'_> {
                         widget.apply_file_search_result(query, matches);
                     }
                 }
+                AppEvent::AttachImage {
+                    path,
+                    width,
+                    height,
+                    format_label,
+                } => {
+                    if let AppState::Chat { widget } = &mut self.app_state {
+                        widget.attach_image(path, width, height, format_label);
+                    }
+                }
             }
         }
         terminal.clear()?;
@@ -649,6 +711,7 @@ mod tests {
     use codex_core::config::ConfigOverrides;
     use codex_core::config::ConfigToml;
     use codex_login::AuthMode;
+    use crossterm::event::KeyModifiers;
 
     fn make_config(preferred: AuthMode) -> Config {
         let mut cfg = Config::load_from_base_config_with_overrides(
@@ -696,4 +759,60 @@ mod tests {
             &cfg
         ))
     }
+
+    #[test]
+    fn ctrl_v_success_attaches_image() {
+        let (tx, rx) = std::sync::mpsc::channel();
+        let sender = AppEventSender::new(tx);
+        let key_event = KeyEvent::new(KeyCode::Char('v'), KeyModifiers::CONTROL);
+        let dummy_info = crate::clipboard_paste::PastedImageInfo {
+            width: 10,
+            height: 5,
+            encoded_format_label: "PNG",
+        };
+        let handled = try_handle_ctrl_v_with(&sender, &key_event, || {
+            Ok((
+                std::path::PathBuf::from("/tmp/test.png"),
+                dummy_info.clone(),
+            ))
+        });
+        assert!(handled, "expected ctrl+v to be handled on success");
+        match rx
+            .recv()
+            .unwrap_or_else(|e| panic!("failed to receive event: {e}"))
+        {
+            AppEvent::AttachImage {
+                path,
+                width,
+                height,
+                format_label,
+            } => {
+                assert_eq!(path, std::path::PathBuf::from("/tmp/test.png"));
+                assert_eq!(width, 10);
+                assert_eq!(height, 5);
+                assert_eq!(format_label, "PNG");
+            }
+            _ => panic!("unexpected event (not AttachImage)"),
+        }
+    }
+
+    #[test]
+    fn ctrl_v_failure_not_consumed() {
+        let (tx, rx) = std::sync::mpsc::channel();
+        let sender = AppEventSender::new(tx);
+        let key_event = KeyEvent::new(KeyCode::Char('v'), KeyModifiers::CONTROL);
+        let handled = try_handle_ctrl_v_with(&sender, &key_event, || {
+            Err(crate::clipboard_paste::PasteImageError::NoImage(
+                "none".into(),
+            ))
+        });
+        assert!(
+            !handled,
+            "on failure ctrl+v should not be considered consumed"
+        );
+        assert!(
+            rx.try_recv().is_err(),
+            "no events should be sent on failure"
+        );
+    }
 }

codex-rs/tui/src/app_event.rs

@@ -7,8 +7,8 @@ use std::time::Duration;
 use crate::app::ChatWidgetArgs;
 use crate::slash_command::SlashCommand;
 
-#[allow(clippy::large_enum_variant)]
 #[derive(Debug)]
+#[allow(clippy::large_enum_variant)]
 pub(crate) enum AppEvent {
     CodexEvent(Event),
 
@@ -63,4 +63,12 @@ pub(crate) enum AppEvent {
     /// Onboarding: result of login_with_chatgpt.
     OnboardingAuthComplete(Result<(), String>),
     OnboardingComplete(ChatWidgetArgs),
+
+    /// Image pasted via Cmd+V (clipboard image attachment).
+    AttachImage {
+        path: std::path::PathBuf,
+        width: u32,
+        height: u32,
+        format_label: &'static str,
+    },
 }

codex-rs/tui/src/bottom_pane/chat_composer.rs

@@ -30,6 +30,7 @@ use crate::bottom_pane::textarea::TextArea;
 use crate::bottom_pane::textarea::TextAreaState;
 use codex_file_search::FileMatch;
 use std::cell::RefCell;
+use std::path::Path;
 
 /// If the pasted content exceeds this number of characters, replace it with a
 /// placeholder in the UI.
@@ -69,6 +70,7 @@ pub(crate) struct ChatComposer {
     pending_pastes: Vec<(String, String)>,
     token_usage_info: Option<TokenUsageInfo>,
     has_focus: bool,
+    attached_images: Vec<(String, std::path::PathBuf)>,
     placeholder_text: String,
 }
 
@@ -101,6 +103,7 @@ impl ChatComposer {
             pending_pastes: Vec::new(),
             token_usage_info: None,
             has_focus: has_input_focus,
+            attached_images: Vec::new(),
             placeholder_text,
         }
     }
@@ -194,6 +197,24 @@ impl ChatComposer {
         true
     }
 
+    pub fn attach_image(
+        &mut self,
+        path: std::path::PathBuf,
+        width: u32,
+        height: u32,
+        format_label: &str,
+    ) -> bool {
+        let placeholder = format!("[image {width}x{height} {format_label}]");
+        self.textarea.insert_str(&placeholder);
+        self.attached_images.push((placeholder, path));
+        true
+    }
+
+    pub fn take_recent_submission_images(&mut self) -> Vec<std::path::PathBuf> {
+        let images = std::mem::take(&mut self.attached_images);
+        images.into_iter().map(|(_, path)| path).collect()
+    }
+
     /// Integrate results from an asynchronous file search.
     pub(crate) fn on_file_search_result(&mut self, query: String, matches: Vec<FileMatch>) {
         // Only apply if user is still editing a token starting with `query`.
@@ -341,12 +362,69 @@ impl ChatComposer {
             } => {
                 if let Some(sel) = popup.selected_match() {
                     let sel_path = sel.to_string();
-                    // Drop popup borrow before using self mutably again.
-                    self.insert_selected_path(&sel_path);
+                    // If selected path looks like an image (png/jpeg), attach as image instead of inserting text.
+                    let is_image = {
+                        let lower = sel_path.to_ascii_lowercase();
+                        lower.ends_with(".png")
+                            || lower.ends_with(".jpg")
+                            || lower.ends_with(".jpeg")
+                    };
+                    if is_image {
+                        // Determine dimensions; if that fails fall back to normal path insertion.
+                        let path_buf = std::path::PathBuf::from(&sel_path);
+                        match image::image_dimensions(&path_buf) {
+                            Ok((w, h)) => {
+                                // Remove the current @token (mirror logic from insert_selected_path without inserting text)
+                                // using the flat text and byte-offset cursor API.
+                                let cursor_offset = self.textarea.cursor();
+                                let text = self.textarea.text();
+                                let before_cursor = &text[..cursor_offset];
+                                let after_cursor = &text[cursor_offset..];
+
+                                // Determine token boundaries in the full text.
+                                let start_idx = before_cursor
+                                    .char_indices()
+                                    .rfind(|(_, c)| c.is_whitespace())
+                                    .map(|(idx, c)| idx + c.len_utf8())
+                                    .unwrap_or(0);
+                                let end_rel_idx = after_cursor
+                                    .char_indices()
+                                    .find(|(_, c)| c.is_whitespace())
+                                    .map(|(idx, _)| idx)
+                                    .unwrap_or(after_cursor.len());
+                                let end_idx = cursor_offset + end_rel_idx;
+
+                                self.textarea.replace_range(start_idx..end_idx, "");
+                                self.textarea.set_cursor(start_idx);
+
+                                let format_label = match Path::new(&sel_path)
+                                    .extension()
+                                    .and_then(|e| e.to_str())
+                                    .map(|s| s.to_ascii_lowercase())
+                                {
+                                    Some(ext) if ext == "png" => "PNG",
+                                    Some(ext) if ext == "jpg" || ext == "jpeg" => "JPEG",
+                                    _ => "IMG",
+                                };
+                                let _ = self.attach_image(path_buf.clone(), w, h, format_label);
+                                // Add a trailing space to keep typing fluid.
+                                self.textarea.insert_str(" ");
+                            }
+                            Err(_) => {
+                                // Fallback to plain path insertion if metadata read fails.
+                                self.insert_selected_path(&sel_path);
+                            }
+                        }
+                    } else {
+                        // Non-image: original behavior.
+                        self.insert_selected_path(&sel_path);
+                    }
                     self.active_popup = ActivePopup::None;
                     return (InputResult::None, true);
                 }
-                (InputResult::None, false)
+                // No selection: treat Enter as closing the popup/session.
+                self.active_popup = ActivePopup::None;
+                (InputResult::None, true)
             }
             input => self.handle_input_basic(input),
         }
@@ -538,12 +616,19 @@ impl ChatComposer {
                 }
                 self.pending_pastes.clear();
 
-                if text.is_empty() {
-                    (InputResult::None, true)
-                } else {
-                    self.history.record_local_submission(&text);
-                    (InputResult::Submitted(text), true)
+                // Strip image placeholders from the submitted text; images are retrieved via take_recent_submission_images()
+                for (placeholder, _) in &self.attached_images {
+                    if text.contains(placeholder) {
+                        text = text.replace(placeholder, "");
+                    }
                 }
+
+                text = text.trim().to_string();
+                if !text.is_empty() {
+                    self.history.record_local_submission(&text);
+                }
+                // Do not clear attached_images here; ChatWidget drains them via take_recent_submission_images().
+                (InputResult::Submitted(text), true)
             }
             input => self.handle_input_basic(input),
         }
@@ -551,6 +636,17 @@ impl ChatComposer {
 
     /// Handle generic Input events that modify the textarea content.
     fn handle_input_basic(&mut self, input: KeyEvent) -> (InputResult, bool) {
+        // Special handling for backspace on placeholders
+        if let KeyEvent {
+            code: KeyCode::Backspace,
+            ..
+        } = input
+        {
+            if self.try_remove_any_placeholder_at_cursor() {
+                return (InputResult::None, true);
+            }
+        }
+
         // Normal input handling
         self.textarea.input(input);
         let text_after = self.textarea.text();
@@ -559,9 +655,61 @@ impl ChatComposer {
         self.pending_pastes
             .retain(|(placeholder, _)| text_after.contains(placeholder));
 
+        // Keep only attached images whose placeholders still exist in text
+        self.attached_images
+            .retain(|(placeholder, _)| text_after.contains(placeholder));
+
         (InputResult::None, true)
     }
 
+    /// Attempts to remove an image or paste placeholder if the cursor is at the end of one.
+    /// Returns true if a placeholder was removed.
+    fn try_remove_any_placeholder_at_cursor(&mut self) -> bool {
+        let p = self.textarea.cursor();
+        let text = self.textarea.text();
+
+        // Try image placeholders first
+        if let Some((idx, placeholder)) =
+            self.attached_images
+                .iter()
+                .enumerate()
+                .find_map(|(i, (ph, _))| {
+                    if p < ph.len() {
+                        return None;
+                    }
+                    let start = p - ph.len();
+                    if text[start..p] == *ph {
+                        Some((i, ph.clone()))
+                    } else {
+                        None
+                    }
+                })
+        {
+            self.textarea.replace_range(p - placeholder.len()..p, "");
+            self.attached_images.remove(idx);
+            return true;
+        }
+
+        // Then try pasted-content placeholders
+        if let Some(placeholder) = self.pending_pastes.iter().find_map(|(ph, _)| {
+            if p < ph.len() {
+                return None;
+            }
+            let start = p - ph.len();
+            if text[start..p] == *ph {
+                Some(ph.clone())
+            } else {
+                None
+            }
+        }) {
+            self.textarea.replace_range(p - placeholder.len()..p, "");
+            self.pending_pastes.retain(|(ph, _)| ph != &placeholder);
+            return true;
+        }
+
+        false
+    }
+
     /// Synchronize `self.command_popup` with the current text in the
     /// textarea. This must be called after every modification that can change
     /// the text so the popup is shown/updated/hidden as appropriate.
@@ -1321,4 +1469,86 @@ mod tests {
             ]
         );
     }
+
+    // --- Image attachment tests ---
+    #[test]
+    fn attach_image_and_submit_includes_image_paths() {
+        use crossterm::event::KeyCode;
+        use crossterm::event::KeyEvent;
+        use crossterm::event::KeyModifiers;
+        let (tx, _rx) = std::sync::mpsc::channel();
+        let sender = AppEventSender::new(tx);
+        let mut composer =
+            ChatComposer::new(true, sender, false, "Ask Codex to do anything".to_string());
+        let path = std::path::PathBuf::from("/tmp/image1.png");
+        assert!(composer.attach_image(path.clone(), 32, 16, "PNG"));
+        composer.handle_paste(" hi".into());
+        let (result, _) =
+            composer.handle_key_event(KeyEvent::new(KeyCode::Enter, KeyModifiers::NONE));
+        match result {
+            InputResult::Submitted(text) => assert_eq!(text, "hi"),
+            _ => panic!("expected Submitted"),
+        }
+        let imgs = composer.take_recent_submission_images();
+        assert_eq!(imgs.len(), 1);
+        assert_eq!(imgs[0], path);
+    }
+
+    #[test]
+    fn attach_image_without_text_submits_empty_text_and_images() {
+        use crossterm::event::KeyCode;
+        use crossterm::event::KeyEvent;
+        use crossterm::event::KeyModifiers;
+        let (tx, _rx) = std::sync::mpsc::channel();
+        let sender = AppEventSender::new(tx);
+        let mut composer =
+            ChatComposer::new(true, sender, false, "Ask Codex to do anything".to_string());
+        let path = std::path::PathBuf::from("/tmp/image2.png");
+        assert!(composer.attach_image(path.clone(), 10, 5, "PNG"));
+        let (result, _) =
+            composer.handle_key_event(KeyEvent::new(KeyCode::Enter, KeyModifiers::NONE));
+        match result {
+            InputResult::Submitted(text) => assert!(text.is_empty()),
+            _ => panic!("expected Submitted"),
+        }
+        let imgs = composer.take_recent_submission_images();
+        assert_eq!(imgs.len(), 1);
+        assert_eq!(imgs[0], path);
+        assert!(composer.attached_images.is_empty());
+    }
+
+    #[test]
+    fn image_placeholder_backspace_behaves_like_text_placeholder() {
+        use crossterm::event::KeyCode;
+        use crossterm::event::KeyEvent;
+        use crossterm::event::KeyModifiers;
+        let (tx, _rx) = std::sync::mpsc::channel();
+        let sender = AppEventSender::new(tx);
+        let mut composer =
+            ChatComposer::new(true, sender, false, "Ask Codex to do anything".to_string());
+        let path = std::path::PathBuf::from("/tmp/image3.png");
+        assert!(composer.attach_image(path.clone(), 20, 10, "PNG"));
+        let placeholder = composer.attached_images[0].0.clone();
+
+        // Case 1: backspace at end
+        composer.textarea.move_cursor_to_end_of_line(false);
+        composer.handle_key_event(KeyEvent::new(KeyCode::Backspace, KeyModifiers::NONE));
+        assert!(!composer.textarea.text().contains(&placeholder));
+        assert!(composer.attached_images.is_empty());
+
+        // Re-add and test backspace in middle: should break the placeholder string
+        // and drop the image mapping (same as text placeholder behavior).
+        assert!(composer.attach_image(path.clone(), 20, 10, "PNG"));
+        let placeholder2 = composer.attached_images[0].0.clone();
+        // Move cursor to roughly middle of placeholder
+        if let Some(start_pos) = composer.textarea.text().find(&placeholder2) {
+            let mid_pos = start_pos + (placeholder2.len() / 2);
+            composer.textarea.set_cursor(mid_pos);
+            composer.handle_key_event(KeyEvent::new(KeyCode::Backspace, KeyModifiers::NONE));
+            assert!(!composer.textarea.text().contains(&placeholder2));
+            assert!(composer.attached_images.is_empty());
+        } else {
+            panic!("Placeholder not found in textarea");
+        }
+    }
 }

codex-rs/tui/src/bottom_pane/mod.rs

@@ -297,6 +297,29 @@ impl BottomPane<'_> {
         self.composer.on_file_search_result(query, matches);
         self.request_redraw();
     }
+
+    pub(crate) fn attach_image(
+        &mut self,
+        path: std::path::PathBuf,
+        width: u32,
+        height: u32,
+        format_label: &str,
+    ) {
+        if self.active_view.is_none() {
+            let needs_redraw = self
+                .composer
+                .attach_image(path, width, height, format_label);
+            if needs_redraw {
+                self.request_redraw();
+            }
+        }
+    }
+
+    pub(crate) fn take_recent_submission_images(&mut self) -> Vec<std::path::PathBuf> {
+        self.composer.take_recent_submission_images()
+    }
+
+    // Removed restart_live_status_with_text – no longer used by the current streaming UI.
 }
 
 impl WidgetRef for &BottomPane<'_> {

codex-rs/tui/src/chatwidget.rs

@@ -543,12 +543,37 @@ impl ChatWidget<'_> {
 
         match self.bottom_pane.handle_key_event(key_event) {
             InputResult::Submitted(text) => {
-                self.submit_user_message(text.into());
+                let images = self.bottom_pane.take_recent_submission_images();
+                self.submit_user_message(UserMessage {
+                    text,
+                    image_paths: images,
+                });
             }
             InputResult::None => {}
         }
     }
 
+    pub(crate) fn attach_image(
+        &mut self,
+        path: std::path::PathBuf,
+        width: u32,
+        height: u32,
+        format_label: &str,
+    ) {
+        tracing::info!(
+            "attach_image path={:?} width={} height={} format={}",
+            path,
+            width,
+            height,
+            format_label
+        );
+        // Forward to bottom pane; width/height/format currently only affect placeholder text.
+        let _ = (width, height, format_label); // reserved for future use (e.g., status flash)
+        self.bottom_pane
+            .attach_image(path.clone(), width, height, format_label);
+        self.request_redraw();
+    }
+
     pub(crate) fn handle_paste(&mut self, text: String) {
         self.bottom_pane.handle_paste(text);
     }

codex-rs/tui/src/clipboard_paste.rs

@@ -0,0 +1,73 @@
+use std::path::PathBuf;
+
+#[derive(Debug)]
+pub enum PasteImageError {
+    ClipboardUnavailable(String),
+    NoImage(String),
+    EncodeFailed(String),
+    IoError(String),
+}
+
+impl std::fmt::Display for PasteImageError {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            PasteImageError::ClipboardUnavailable(msg) => write!(f, "clipboard unavailable: {msg}"),
+            PasteImageError::NoImage(msg) => write!(f, "no image on clipboard: {msg}"),
+            PasteImageError::EncodeFailed(msg) => write!(f, "could not encode image: {msg}"),
+            PasteImageError::IoError(msg) => write!(f, "io error: {msg}"),
+        }
+    }
+}
+impl std::error::Error for PasteImageError {}
+
+#[derive(Debug, Clone)]
+pub struct PastedImageInfo {
+    pub width: u32,
+    pub height: u32,
+    pub encoded_format_label: &'static str, // Always PNG for now.
+}
+
+/// Capture image from system clipboard, encode to PNG, and return bytes + info.
+pub fn paste_image_as_png() -> Result<(Vec<u8>, PastedImageInfo), PasteImageError> {
+    tracing::debug!("attempting clipboard image read");
+    let mut cb = arboard::Clipboard::new()
+        .map_err(|e| PasteImageError::ClipboardUnavailable(e.to_string()))?;
+    let img = cb
+        .get_image()
+        .map_err(|e| PasteImageError::NoImage(e.to_string()))?;
+    let w = img.width as u32;
+    let h = img.height as u32;
+
+    let mut png: Vec<u8> = Vec::new();
+    let Some(rgba_img) = image::RgbaImage::from_raw(w, h, img.bytes.into_owned()) else {
+        return Err(PasteImageError::EncodeFailed("invalid RGBA buffer".into()));
+    };
+    let dyn_img = image::DynamicImage::ImageRgba8(rgba_img);
+    tracing::debug!("clipboard image decoded RGBA {}x{}", w, h);
+    {
+        let mut cursor = std::io::Cursor::new(&mut png);
+        dyn_img
+            .write_to(&mut cursor, image::ImageFormat::Png)
+            .map_err(|e| PasteImageError::EncodeFailed(e.to_string()))?;
+    }
+
+    tracing::debug!("clipboard image encoded to PNG ({} bytes)", png.len());
+    Ok((
+        png,
+        PastedImageInfo {
+            width: w,
+            height: h,
+            encoded_format_label: "PNG",
+        },
+    ))
+}
+
+/// Convenience: write to a temp file and return its path + info.
+pub fn paste_image_to_temp_png() -> Result<(PathBuf, PastedImageInfo), PasteImageError> {
+    let (png, info) = paste_image_as_png()?;
+    let mut path = std::env::temp_dir();
+    let fname = format!("clipboard-{}x{}.png", info.width, info.height);
+    path.push(fname);
+    std::fs::write(&path, &png).map_err(|e| PasteImageError::IoError(e.to_string()))?;
+    Ok((path, info))
+}

codex-rs/tui/src/lib.rs

@@ -30,6 +30,7 @@ mod bottom_pane;
 mod chatwidget;
 mod citation_regex;
 mod cli;
+mod clipboard_paste;
 mod common;
 pub mod custom_terminal;
 mod diff_render;