image-gen-core (#13290)

Core tool-calling for image-gen, handles requesting and receiving logic
for images using response API

codex-rs/core/src/context_manager/history.rs

@@ -344,6 +344,9 @@ impl ContextManager {
         // all outputs must have a corresponding function/tool call
         normalize::remove_orphan_outputs(&mut self.items);
 
+        //rewrite image_gen_calls to messages to support stateless input
+        normalize::rewrite_image_generation_calls_for_stateless_input(&mut self.items);
+
         // strip images when model does not support them
         normalize::strip_images_when_unsupported(input_modalities, &mut self.items);
     }
@@ -374,6 +377,7 @@ impl ContextManager {
             | ResponseItem::LocalShellCall { .. }
             | ResponseItem::FunctionCall { .. }
             | ResponseItem::WebSearchCall { .. }
+            | ResponseItem::ImageGenerationCall { .. }
             | ResponseItem::CustomToolCall { .. }
             | ResponseItem::Compaction { .. }
             | ResponseItem::GhostSnapshot { .. }
@@ -402,7 +406,8 @@ fn truncate_function_output_payload(
 }
 
 /// API messages include every non-system item (user/assistant messages, reasoning,
-/// tool calls, tool outputs, shell calls, and web-search calls).
+/// tool calls, tool outputs, shell calls, web-search calls, and image-generation
+/// calls).
 fn is_api_message(message: &ResponseItem) -> bool {
     match message {
         ResponseItem::Message { role, .. } => role.as_str() != "system",
@@ -413,6 +418,7 @@ fn is_api_message(message: &ResponseItem) -> bool {
         | ResponseItem::LocalShellCall { .. }
         | ResponseItem::Reasoning { .. }
         | ResponseItem::WebSearchCall { .. }
+        | ResponseItem::ImageGenerationCall { .. }
         | ResponseItem::Compaction { .. } => true,
         ResponseItem::GhostSnapshot { .. } => false,
         ResponseItem::Other => false,
@@ -600,6 +606,7 @@ fn is_model_generated_item(item: &ResponseItem) -> bool {
         ResponseItem::Reasoning { .. }
         | ResponseItem::FunctionCall { .. }
         | ResponseItem::WebSearchCall { .. }
+        | ResponseItem::ImageGenerationCall { .. }
         | ResponseItem::CustomToolCall { .. }
         | ResponseItem::LocalShellCall { .. }
         | ResponseItem::Compaction { .. } => true,

codex-rs/core/src/context_manager/history_tests.rs

@@ -395,6 +395,97 @@ fn for_prompt_strips_images_when_model_does_not_support_images() {
     }
 }
 
+#[test]
+fn for_prompt_rewrites_image_generation_calls_when_images_are_supported() {
+    let history = create_history_with_items(vec![
+        ResponseItem::ImageGenerationCall {
+            id: "ig_123".to_string(),
+            status: "generating".to_string(),
+            revised_prompt: Some("lobster".to_string()),
+            result: "Zm9v".to_string(),
+        },
+        ResponseItem::Message {
+            id: None,
+            role: "user".to_string(),
+            content: vec![ContentItem::InputText {
+                text: "hi".to_string(),
+            }],
+            end_turn: None,
+            phase: None,
+        },
+    ]);
+
+    assert_eq!(
+        history.for_prompt(&default_input_modalities()),
+        vec![
+            ResponseItem::Message {
+                id: None,
+                role: "user".to_string(),
+                content: vec![ContentItem::InputImage {
+                    image_url: "data:image/png;base64,Zm9v".to_string(),
+                }],
+                end_turn: None,
+                phase: None,
+            },
+            ResponseItem::Message {
+                id: None,
+                role: "user".to_string(),
+                content: vec![ContentItem::InputText {
+                    text: "hi".to_string(),
+                }],
+                end_turn: None,
+                phase: None,
+            }
+        ]
+    );
+}
+
+#[test]
+fn for_prompt_rewrites_image_generation_calls_when_images_are_unsupported() {
+    let history = create_history_with_items(vec![
+        ResponseItem::Message {
+            id: None,
+            role: "user".to_string(),
+            content: vec![ContentItem::InputText {
+                text: "generate a lobster".to_string(),
+            }],
+            end_turn: None,
+            phase: None,
+        },
+        ResponseItem::ImageGenerationCall {
+            id: "ig_123".to_string(),
+            status: "completed".to_string(),
+            revised_prompt: Some("lobster".to_string()),
+            result: "Zm9v".to_string(),
+        },
+    ]);
+
+    assert_eq!(
+        history.for_prompt(&[InputModality::Text]),
+        vec![
+            ResponseItem::Message {
+                id: None,
+                role: "user".to_string(),
+                content: vec![ContentItem::InputText {
+                    text: "generate a lobster".to_string(),
+                }],
+                end_turn: None,
+                phase: None,
+            },
+            ResponseItem::Message {
+                id: None,
+                role: "user".to_string(),
+                content: vec![ContentItem::InputText {
+                    text: "image content omitted because you do not support image input"
+                        .to_string(),
+                }],
+                end_turn: None,
+                phase: None,
+            },
+        ]
+    );
+}
+
 #[test]
 fn get_history_for_prompt_drops_ghost_commits() {
     let items = vec![ResponseItem::GhostSnapshot {

codex-rs/core/src/context_manager/normalize.rs

@@ -1,10 +1,9 @@
-use std::collections::HashSet;
-
 use codex_protocol::models::ContentItem;
 use codex_protocol::models::FunctionCallOutputContentItem;
 use codex_protocol::models::FunctionCallOutputPayload;
 use codex_protocol::models::ResponseItem;
 use codex_protocol::openai_models::InputModality;
+use std::collections::HashSet;
 
 use crate::util::error_or_panic;
 use tracing::info;
@@ -211,6 +210,31 @@ where
     }
 }
 
+pub(crate) fn rewrite_image_generation_calls_for_stateless_input(items: &mut Vec<ResponseItem>) {
+    let original_items = std::mem::take(items);
+    *items = original_items
+        .into_iter()
+        .map(|item| match item {
+            ResponseItem::ImageGenerationCall { result, .. } => {
+                let image_url = if result.starts_with("data:") {
+                    result
+                } else {
+                    format!("data:image/png;base64,{result}")
+                };
+
+                ResponseItem::Message {
+                    id: None,
+                    role: "user".to_string(),
+                    content: vec![ContentItem::InputImage { image_url }],
+                    end_turn: None,
+                    phase: None,
+                }
+            }
+            _ => item,
+        })
+        .collect();
+}
+
 /// Strip image content from messages and tool outputs when the model does not support images.
 /// When `input_modalities` contains `InputModality::Image`, no stripping is performed.
 pub(crate) fn strip_images_when_unsupported(