feat(app-server, core): allow text + image content items for dynamic tool outputs (#10567)

Took over the work that @aaronl-openai started here:
https://github.com/openai/codex/pull/10397

Now that app-server clients are able to set up custom tools (called
`dynamic_tools` in app-server), we should expose a way for clients to
pass in not just text, but also image outputs. This is something the
Responses API already supports for function call outputs, where you can
pass in either a string or an array of content outputs (text, image,
file):
https://platform.openai.com/docs/api-reference/responses/create#responses_create-input-input_item_list-item-function_tool_call_output-output-array-input_image

So let's just plumb it through in Codex (with the caveat that we only
support text and image for now). This is implemented end-to-end across
app-server v2 protocol types and core tool handling.

## Breaking API change
NOTE: This introduces a breaking change with dynamic tools, but I think
it's ok since this concept was only recently introduced
(https://github.com/openai/codex/pull/9539) and it's better to get the
API contract correct. I don't think there are any real consumers of this
yet (not even the Codex App).

Old shape:
`{ "output": "dynamic-ok", "success": true }`

New shape:
```
{
    "contentItems": [
      { "type": "inputText", "text": "dynamic-ok" },
      { "type": "inputImage", "imageUrl": "data:image/png;base64,AAA" }
    ]
  "success": true
}
```

codex-rs/app-server-protocol/schema/json/ClientRequest.json

@@ -480,6 +480,19 @@
       },
       "type": "object"
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -526,19 +539,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -548,7 +552,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/DynamicToolCallResponse.json

@@ -1,15 +1,64 @@
 {
   "$schema": "http://json-schema.org/draft-07/schema#",
+  "definitions": {
+    "DynamicToolCallOutputContentItem": {
+      "oneOf": [
+        {
+          "properties": {
+            "text": {
+              "type": "string"
+            },
+            "type": {
+              "enum": [
+                "inputText"
+              ],
+              "title": "InputTextDynamicToolCallOutputContentItemType",
+              "type": "string"
+            }
+          },
+          "required": [
+            "text",
+            "type"
+          ],
+          "title": "InputTextDynamicToolCallOutputContentItem",
+          "type": "object"
+        },
+        {
+          "properties": {
+            "imageUrl": {
+              "type": "string"
+            },
+            "type": {
+              "enum": [
+                "inputImage"
+              ],
+              "title": "InputImageDynamicToolCallOutputContentItemType",
+              "type": "string"
+            }
+          },
+          "required": [
+            "imageUrl",
+            "type"
+          ],
+          "title": "InputImageDynamicToolCallOutputContentItem",
+          "type": "object"
+        }
+      ]
+    }
+  },
   "properties": {
-    "output": {
-      "type": "string"
+    "contentItems": {
+      "items": {
+        "$ref": "#/definitions/DynamicToolCallOutputContentItem"
+      },
+      "type": "array"
     },
     "success": {
       "type": "boolean"
     }
   },
   "required": [
-    "output",
+    "contentItems",
     "success"
   ],
   "title": "DynamicToolCallResponse",

codex-rs/app-server-protocol/schema/json/EventMsg.json

@@ -2864,6 +2864,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -2910,19 +2923,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -2932,7 +2936,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/ServerNotification.json

@@ -3484,6 +3484,19 @@
       ],
       "type": "object"
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -3530,19 +3543,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -3552,7 +3556,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/codex_app_server_protocol.schemas.json

@@ -2312,6 +2312,50 @@
       ],
       "type": "object"
     },
+    "DynamicToolCallOutputContentItem": {
+      "oneOf": [
+        {
+          "properties": {
+            "text": {
+              "type": "string"
+            },
+            "type": {
+              "enum": [
+                "inputText"
+              ],
+              "title": "InputTextDynamicToolCallOutputContentItemType",
+              "type": "string"
+            }
+          },
+          "required": [
+            "text",
+            "type"
+          ],
+          "title": "InputTextDynamicToolCallOutputContentItem",
+          "type": "object"
+        },
+        {
+          "properties": {
+            "imageUrl": {
+              "type": "string"
+            },
+            "type": {
+              "enum": [
+                "inputImage"
+              ],
+              "title": "InputImageDynamicToolCallOutputContentItemType",
+              "type": "string"
+            }
+          },
+          "required": [
+            "imageUrl",
+            "type"
+          ],
+          "title": "InputImageDynamicToolCallOutputContentItem",
+          "type": "object"
+        }
+      ]
+    },
     "DynamicToolCallParams": {
       "$schema": "http://json-schema.org/draft-07/schema#",
       "properties": {
@@ -2342,15 +2386,18 @@
     "DynamicToolCallResponse": {
       "$schema": "http://json-schema.org/draft-07/schema#",
       "properties": {
-        "output": {
-          "type": "string"
+        "contentItems": {
+          "items": {
+            "$ref": "#/definitions/DynamicToolCallOutputContentItem"
+          },
+          "type": "array"
         },
         "success": {
           "type": "boolean"
         }
       },
       "required": [
-        "output",
+        "contentItems",
         "success"
       ],
       "title": "DynamicToolCallResponse",
@@ -5044,6 +5091,19 @@
       "title": "ForkConversationResponse",
       "type": "object"
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -5090,19 +5150,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -5112,7 +5163,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },
@@ -11246,6 +11297,19 @@
         ],
         "type": "string"
       },
+      "FunctionCallOutputBody": {
+        "anyOf": [
+          {
+            "type": "string"
+          },
+          {
+            "items": {
+              "$ref": "#/definitions/v2/FunctionCallOutputContentItem"
+            },
+            "type": "array"
+          }
+        ]
+      },
       "FunctionCallOutputContentItem": {
         "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
         "oneOf": [
@@ -11292,19 +11356,10 @@
         ]
       },
       "FunctionCallOutputPayload": {
-        "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+        "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
         "properties": {
-          "content": {
-            "type": "string"
-          },
-          "content_items": {
-            "items": {
-              "$ref": "#/definitions/v2/FunctionCallOutputContentItem"
-            },
-            "type": [
-              "array",
-              "null"
-            ]
+          "body": {
+            "$ref": "#/definitions/v2/FunctionCallOutputBody"
           },
           "success": {
             "type": [
@@ -11314,7 +11369,7 @@
           }
         },
         "required": [
-          "content"
+          "body"
         ],
         "type": "object"
       },

codex-rs/app-server-protocol/schema/json/v1/ForkConversationResponse.json

@@ -2864,6 +2864,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -2910,19 +2923,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -2932,7 +2936,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/v1/ResumeConversationParams.json

@@ -98,6 +98,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -144,19 +157,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -166,7 +170,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/v1/ResumeConversationResponse.json

@@ -2864,6 +2864,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -2910,19 +2923,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -2932,7 +2936,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/v1/SessionConfiguredNotification.json

@@ -2864,6 +2864,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -2910,19 +2923,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -2932,7 +2936,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/v2/RawResponseItemCompletedNotification.json

@@ -65,6 +65,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -111,19 +124,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -133,7 +137,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/json/v2/ThreadResumeParams.json

@@ -74,6 +74,19 @@
         }
       ]
     },
+    "FunctionCallOutputBody": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "items": {
+            "$ref": "#/definitions/FunctionCallOutputContentItem"
+          },
+          "type": "array"
+        }
+      ]
+    },
     "FunctionCallOutputContentItem": {
       "description": "Responses API compatible content items that can be returned by a tool call. This is a subset of ContentItem with the types we support as function call outputs.",
       "oneOf": [
@@ -120,19 +133,10 @@
       ]
     },
     "FunctionCallOutputPayload": {
-      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`content` preserves the historical plain-string payload so downstream integrations (tests, logging, etc.) can keep treating tool output as `String`. When an MCP server returns richer data we additionally populate `content_items` with the structured form that the Responses API understands.",
+      "description": "The payload we send back to OpenAI when reporting a tool call result.\n\n`body` serializes directly as the wire value for `function_call_output.output`. `success` remains internal metadata for downstream handling.",
       "properties": {
-        "content": {
-          "type": "string"
-        },
-        "content_items": {
-          "items": {
-            "$ref": "#/definitions/FunctionCallOutputContentItem"
-          },
-          "type": [
-            "array",
-            "null"
-          ]
+        "body": {
+          "$ref": "#/definitions/FunctionCallOutputBody"
         },
         "success": {
           "type": [
@@ -142,7 +146,7 @@
         }
       },
       "required": [
-        "content"
+        "body"
       ],
       "type": "object"
     },

codex-rs/app-server-protocol/schema/typescript/FunctionCallOutputBody.ts

@@ -0,0 +1,6 @@
+// GENERATED CODE! DO NOT MODIFY BY HAND!
+
+// This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
+import type { FunctionCallOutputContentItem } from "./FunctionCallOutputContentItem";
+
+export type FunctionCallOutputBody = string | Array<FunctionCallOutputContentItem>;

codex-rs/app-server-protocol/schema/typescript/FunctionCallOutputPayload.ts

@@ -1,14 +1,12 @@
 // GENERATED CODE! DO NOT MODIFY BY HAND!
 
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
-import type { FunctionCallOutputContentItem } from "./FunctionCallOutputContentItem";
+import type { FunctionCallOutputBody } from "./FunctionCallOutputBody";
 
 /**
  * The payload we send back to OpenAI when reporting a tool call result.
  *
- * `content` preserves the historical plain-string payload so downstream
- * integrations (tests, logging, etc.) can keep treating tool output as
- * `String`. When an MCP server returns richer data we additionally populate
- * `content_items` with the structured form that the Responses API understands.
+ * `body` serializes directly as the wire value for `function_call_output.output`.
+ * `success` remains internal metadata for downstream handling.
  */
-export type FunctionCallOutputPayload = { content: string, content_items: Array<FunctionCallOutputContentItem> | null, success: boolean | null, };
+export type FunctionCallOutputPayload = { body: FunctionCallOutputBody, success: boolean | null, };

codex-rs/app-server-protocol/schema/typescript/index.ts

@@ -69,6 +69,7 @@ export type { FileChange } from "./FileChange";
 export type { ForcedLoginMethod } from "./ForcedLoginMethod";
 export type { ForkConversationParams } from "./ForkConversationParams";
 export type { ForkConversationResponse } from "./ForkConversationResponse";
+export type { FunctionCallOutputBody } from "./FunctionCallOutputBody";
 export type { FunctionCallOutputContentItem } from "./FunctionCallOutputContentItem";
 export type { FunctionCallOutputPayload } from "./FunctionCallOutputPayload";
 export type { FuzzyFileSearchParams } from "./FuzzyFileSearchParams";

codex-rs/app-server-protocol/schema/typescript/v2/DynamicToolCallOutputContentItem.ts

@@ -0,0 +1,5 @@
+// GENERATED CODE! DO NOT MODIFY BY HAND!
+
+// This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
+
+export type DynamicToolCallOutputContentItem = { "type": "inputText", text: string, } | { "type": "inputImage", imageUrl: string, };

codex-rs/app-server-protocol/schema/typescript/v2/DynamicToolCallResponse.ts

@@ -1,5 +1,6 @@
 // GENERATED CODE! DO NOT MODIFY BY HAND!
 
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
+import type { DynamicToolCallOutputContentItem } from "./DynamicToolCallOutputContentItem";
 
-export type DynamicToolCallResponse = { output: string, success: boolean, };
+export type DynamicToolCallResponse = { contentItems: Array<DynamicToolCallOutputContentItem>, success: boolean, };

codex-rs/app-server-protocol/schema/typescript/v2/index.ts

@@ -46,6 +46,7 @@ export type { ConfigWriteResponse } from "./ConfigWriteResponse";
 export type { ContextCompactedNotification } from "./ContextCompactedNotification";
 export type { CreditsSnapshot } from "./CreditsSnapshot";
 export type { DeprecationNoticeNotification } from "./DeprecationNoticeNotification";
+export type { DynamicToolCallOutputContentItem } from "./DynamicToolCallOutputContentItem";
 export type { DynamicToolCallParams } from "./DynamicToolCallParams";
 export type { DynamicToolCallResponse } from "./DynamicToolCallResponse";
 export type { DynamicToolSpec } from "./DynamicToolSpec";

codex-rs/app-server-protocol/src/protocol/v2.rs

@@ -2826,10 +2826,34 @@ pub struct DynamicToolCallParams {
 #[serde(rename_all = "camelCase")]
 #[ts(export_to = "v2/")]
 pub struct DynamicToolCallResponse {
-    pub output: String,
+    pub content_items: Vec<DynamicToolCallOutputContentItem>,
     pub success: bool,
 }
 
+#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, JsonSchema, TS)]
+#[serde(tag = "type", rename_all = "camelCase")]
+#[ts(tag = "type")]
+#[ts(export_to = "v2/")]
+pub enum DynamicToolCallOutputContentItem {
+    #[serde(rename_all = "camelCase")]
+    InputText { text: String },
+    #[serde(rename_all = "camelCase")]
+    InputImage { image_url: String },
+}
+
+impl From<DynamicToolCallOutputContentItem>
+    for codex_protocol::dynamic_tools::DynamicToolCallOutputContentItem
+{
+    fn from(item: DynamicToolCallOutputContentItem) -> Self {
+        match item {
+            DynamicToolCallOutputContentItem::InputText { text } => Self::InputText { text },
+            DynamicToolCallOutputContentItem::InputImage { image_url } => {
+                Self::InputImage { image_url }
+            }
+        }
+    }
+}
+
 #[derive(Serialize, Deserialize, Debug, Clone, PartialEq, JsonSchema, TS)]
 #[serde(rename_all = "camelCase")]
 #[ts(export_to = "v2/")]
@@ -3189,4 +3213,61 @@ mod tests {
             })
         );
     }
+
+    #[test]
+    fn dynamic_tool_response_serializes_content_items() {
+        let value = serde_json::to_value(DynamicToolCallResponse {
+            content_items: vec![DynamicToolCallOutputContentItem::InputText {
+                text: "dynamic-ok".to_string(),
+            }],
+            success: true,
+        })
+        .unwrap();
+
+        assert_eq!(
+            value,
+            json!({
+                "contentItems": [
+                    {
+                        "type": "inputText",
+                        "text": "dynamic-ok"
+                    }
+                ],
+                "success": true,
+            })
+        );
+    }
+
+    #[test]
+    fn dynamic_tool_response_serializes_text_and_image_content_items() {
+        let value = serde_json::to_value(DynamicToolCallResponse {
+            content_items: vec![
+                DynamicToolCallOutputContentItem::InputText {
+                    text: "dynamic-ok".to_string(),
+                },
+                DynamicToolCallOutputContentItem::InputImage {
+                    image_url: "data:image/png;base64,AAA".to_string(),
+                },
+            ],
+            success: true,
+        })
+        .unwrap();
+
+        assert_eq!(
+            value,
+            json!({
+                "contentItems": [
+                    {
+                        "type": "inputText",
+                        "text": "dynamic-ok"
+                    },
+                    {
+                        "type": "inputImage",
+                        "imageUrl": "data:image/png;base64,AAA"
+                    }
+                ],
+                "success": true,
+            })
+        );
+    }
 }