@rdmind/rdmind 0.2.8-alpha.0 → 0.2.8-alpha.2

package/cli.js

@@ -147834,6 +147834,7 @@ var init_vertexAnthropicContentGenerator = __esm({
     "use strict";
     init_esbuild_shims();
     init_node();
+    init_modalityDefaults();
     init_errorHandler();
     init_loggers();
     init_types();
@@ -147907,6 +147908,7 @@ var init_vertexAnthropicContentGenerator = __esm({
       apiKey;
       samplingParams;
       reasoning;
+      modalities;
       cliConfig;
       telemetryService;
       errorHandler;
@@ -147915,6 +147917,7 @@ var init_vertexAnthropicContentGenerator = __esm({
         this.apiKey = config2.apiKey || "";
         this.samplingParams = config2.samplingParams;
         this.reasoning = config2.reasoning;
+        this.modalities = config2.modalities ?? defaultModalities(config2.model);
         this.cliConfig = cliConfig;
         if (!this.apiKey) {
           throw new Error("API key is required for Vertex Anthropic");
@@ -148093,7 +148096,14 @@ var init_vertexAnthropicContentGenerator = __esm({
         }
         if (part.inlineData?.mimeType && part.inlineData?.data) {
           const mimeType = part.inlineData.mimeType;
+          const displayName = part.inlineData.displayName || mimeType;
           if (this.isSupportedImageMimeType(mimeType)) {
+            if (!this.modalities.image) {
+              return {
+                type: "text",
+                text: this.unsupportedModalityHint("image", displayName)
+              };
+            }
             return {
               type: "image",
               source: {
@@ -148104,6 +148114,12 @@ var init_vertexAnthropicContentGenerator = __esm({
             };
           }
           if (mimeType === "application/pdf") {
+            if (!this.modalities.pdf) {
+              return {
+                type: "text",
+                text: this.unsupportedModalityHint("pdf", displayName)
+              };
+            }
             return {
               type: "document",
               source: {
@@ -148113,13 +148129,54 @@ var init_vertexAnthropicContentGenerator = __esm({
               }
             };
           }
-          const displayName = part.inlineData.displayName ? ` (${part.inlineData.displayName})` : "";
+          if (mimeType.startsWith("audio/")) {
+            if (!this.modalities.audio) {
+              return {
+                type: "text",
+                text: this.unsupportedModalityHint("audio", displayName)
+              };
+            }
+          }
+          if (mimeType.startsWith("video/")) {
+            if (!this.modalities.video) {
+              return {
+                type: "text",
+                text: this.unsupportedModalityHint("video", displayName)
+              };
+            }
+          }
           return {
             type: "text",
-            text: `[Unsupported media type: ${mimeType}${displayName}]`
+            text: `[Unsupported media type: ${mimeType} (${displayName})]`
           };
         }
         if (part.fileData?.mimeType && part.fileData?.fileUri) {
+          const fileDisplayName = part.fileData.displayName || part.fileData.fileUri;
+          const fileMimeType = part.fileData.mimeType;
+          if (fileMimeType.startsWith("image/") && !this.modalities.image) {
+            return {
+              type: "text",
+              text: this.unsupportedModalityHint("image", fileDisplayName)
+            };
+          }
+          if (fileMimeType === "application/pdf" && !this.modalities.pdf) {
+            return {
+              type: "text",
+              text: this.unsupportedModalityHint("pdf", fileDisplayName)
+            };
+          }
+          if (fileMimeType.startsWith("audio/") && !this.modalities.audio) {
+            return {
+              type: "text",
+              text: this.unsupportedModalityHint("audio", fileDisplayName)
+            };
+          }
+          if (fileMimeType.startsWith("video/") && !this.modalities.video) {
+            return {
+              type: "text",
+              text: this.unsupportedModalityHint("video", fileDisplayName)
+            };
+          }
           return {
             type: "text",
             text: `[External file reference: ${part.fileData.fileUri}]`
@@ -148155,6 +148212,13 @@ var init_vertexAnthropicContentGenerator = __esm({
       isSupportedImageMimeType(mimeType) {
         return mimeType === "image/jpeg" || mimeType === "image/png" || mimeType === "image/gif" || mimeType === "image/webp";
       }
+      unsupportedModalityHint(modality, displayName) {
+        if (modality === "pdf") {
+          return `[Unsupported pdf file: "${displayName}". This model does not support PDF input directly. The read_file tool cannot extract PDF content either. To extract text from the PDF file, try using skills if applicable, or guide user to install pdf skill by running this slash command:
+/extensions install https://github.com/anthropics/skills:document-skills]`;
+        }
+        return `[Unsupported ${modality} file: "${displayName}". This model does not support ${modality} input. The read_file tool cannot process this type of file either. To handle this file, try using skills if applicable, or any tools installed at system wide, or let the user know you cannot process this type of file.]`;
+      }
       /**
        * 将 Gemini 工具定义转换为 Anthropic 格式
        * 参考 AnthropicContentConverter.convertGeminiToolsToAnthropic
@@ -148678,7 +148742,7 @@ function extractReasoningEffort(model) {
   }
   return void 0;
 }
-function buildCodexInput(contents, systemInstruction) {
+function buildCodexInput(contents, systemInstruction, modalities) {
   const messages = [];
   if (systemInstruction) {
     const text = extractTextFromContent(systemInstruction);
@@ -148688,7 +148752,7 @@ function buildCodexInput(contents, systemInstruction) {
   }
   const contentsArray = Array.isArray(contents) ? contents : [contents];
   for (const content of contentsArray) {
-    const codexMessages = convertContentToCodexMessages(content);
+    const codexMessages = convertContentToCodexMessages(content, modalities);
     messages.push(...codexMessages);
   }
   if (messages.length === 1 && messages[0].role === "user") {
@@ -148699,7 +148763,7 @@ function buildCodexInput(contents, systemInstruction) {
   }
   return messages;
 }
-function convertContentToCodexMessages(content) {
+function convertContentToCodexMessages(content, modalities) {
   const messages = [];
   if (!content) return messages;
   if (typeof content === "string") {
@@ -148742,14 +148806,14 @@ function convertContentToCodexMessages(content) {
       }
       if ("inlineData" in partObj && partObj["inlineData"]) {
         const inlineData = partObj["inlineData"];
-        const mediaMsg = convertInlineDataToCodexMessage(inlineData);
+        const mediaMsg = convertInlineDataToCodexMessage(inlineData, modalities);
         if (mediaMsg) {
           messages.push(mediaMsg);
         }
       }
       if ("fileData" in partObj && partObj["fileData"]) {
         const fileData = partObj["fileData"];
-        const mediaMsg = convertFileDataToCodexMessage(fileData);
+        const mediaMsg = convertFileDataToCodexMessage(fileData, modalities);
         if (mediaMsg) {
           messages.push(mediaMsg);
         }
@@ -148758,10 +148822,18 @@ function convertContentToCodexMessages(content) {
   }
   return messages;
 }
-function convertInlineDataToCodexMessage(inlineData) {
+function convertInlineDataToCodexMessage(inlineData, modalities) {
   const mimeType = inlineData.mimeType;
   const dataUrl = `data:${mimeType};base64,${inlineData.data}`;
+  const displayName = inlineData.displayName || mimeType;
   if (mimeType.startsWith("image/")) {
+    if (!modalities.image) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("image", displayName)
+      };
+    }
     return {
       role: "user",
       type: "message",
@@ -148769,6 +148841,13 @@ function convertInlineDataToCodexMessage(inlineData) {
     };
   }
   if (mimeType === "application/pdf") {
+    if (!modalities.pdf) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("pdf", displayName)
+      };
+    }
     return {
       role: "user",
       type: "message",
@@ -148776,22 +148855,53 @@ function convertInlineDataToCodexMessage(inlineData) {
     };
   }
   if (mimeType.startsWith("audio/")) {
+    if (!modalities.audio) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("audio", displayName)
+      };
+    }
     return {
       role: "user",
       type: "message",
       content: `[Audio file: ${inlineData.displayName || "audio"} (${mimeType})]`
     };
   }
+  if (mimeType.startsWith("video/")) {
+    if (!modalities.video) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("video", displayName)
+      };
+    }
+  }
   return {
     role: "user",
     type: "message",
     content: `[Unsupported file: ${mimeType}${inlineData.displayName ? ` (${inlineData.displayName})` : ""}]`
   };
 }
-function convertFileDataToCodexMessage(fileData) {
+function unsupportedModalityHint(modality, displayName) {
+  if (modality === "pdf") {
+    return `[Unsupported pdf file: "${displayName}". This model does not support PDF input directly. The read_file tool cannot extract PDF content either. To extract text from the PDF file, try using skills if applicable, or guide user to install pdf skill by running this slash command:
+/extensions install https://github.com/anthropics/skills:document-skills]`;
+  }
+  return `[Unsupported ${modality} file: "${displayName}". This model does not support ${modality} input. The read_file tool cannot process this type of file either. To handle this file, try using skills if applicable, or any tools installed at system wide, or let the user know you cannot process this type of file.]`;
+}
+function convertFileDataToCodexMessage(fileData, modalities) {
   const mimeType = fileData.mimeType;
   const fileUri = fileData.fileUri;
+  const displayName = fileData.displayName || mimeType;
   if (mimeType.startsWith("image/")) {
+    if (!modalities.image) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("image", displayName)
+      };
+    }
     return {
       role: "user",
       type: "message",
@@ -148799,6 +148909,13 @@ function convertFileDataToCodexMessage(fileData) {
     };
   }
   if (mimeType === "application/pdf") {
+    if (!modalities.pdf) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("pdf", displayName)
+      };
+    }
     return {
       role: "user",
       type: "message",
@@ -148806,12 +148923,28 @@ function convertFileDataToCodexMessage(fileData) {
     };
   }
   if (mimeType.startsWith("audio/")) {
+    if (!modalities.audio) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("audio", displayName)
+      };
+    }
     return {
       role: "user",
       type: "message",
       content: `[Audio file: ${fileData.displayName || "audio"} (${mimeType})]`
     };
   }
+  if (mimeType.startsWith("video/")) {
+    if (!modalities.video) {
+      return {
+        role: "user",
+        type: "message",
+        content: unsupportedModalityHint("video", displayName)
+      };
+    }
+  }
   return {
     role: "user",
     type: "message",
@@ -148954,6 +149087,7 @@ var init_codexContentGenerator = __esm({
     "use strict";
     init_esbuild_shims();
     init_node();
+    init_modalityDefaults();
     init_errorHandler();
     init_loggers();
     init_types();
@@ -148966,6 +149100,7 @@ var init_codexContentGenerator = __esm({
       apiKey;
       samplingParams;
       reasoning;
+      modalities;
       cliConfig;
       errorHandler;
       logger;
@@ -148974,6 +149109,7 @@ var init_codexContentGenerator = __esm({
         this.apiKey = config2.apiKey || "";
         this.samplingParams = config2.samplingParams;
         this.reasoning = config2.reasoning;
+        this.modalities = config2.modalities ?? defaultModalities(config2.model);
         this.cliConfig = cliConfig;
         if (!this.apiKey) {
           throw new Error("API key is required for Codex");
@@ -149064,7 +149200,8 @@ var init_codexContentGenerator = __esm({
         const reasoningEffort = extractReasoningEffort(request4.model) || (typeof this.reasoning === "object" ? this.reasoning?.effort : "medium");
         const input = buildCodexInput(
           request4.contents,
-          request4.config?.systemInstruction
+          request4.config?.systemInstruction,
+          this.modalities
         );
         const codexRequest = {
           model,
@@ -149249,6 +149386,7 @@ var init_codexContentGenerator = __esm({
     __name(buildCodexInput, "buildCodexInput");
     __name(convertContentToCodexMessages, "convertContentToCodexMessages");
     __name(convertInlineDataToCodexMessage, "convertInlineDataToCodexMessage");
+    __name(unsupportedModalityHint, "unsupportedModalityHint");
     __name(convertFileDataToCodexMessage, "convertFileDataToCodexMessage");
     __name(extractTextFromContent, "extractTextFromContent");
     __name(convertTools, "convertTools");
@@ -160844,7 +160982,7 @@ __export(geminiContentGenerator_exports2, {
   createGeminiContentGenerator: () => createGeminiContentGenerator
 });
 function createGeminiContentGenerator(config2, gcConfig) {
-  const version2 = "0.2.8-alpha.0";
+  const version2 = "0.2.8-alpha.2";
   const userAgent2 = config2.userAgent || `QwenCode/${version2} (${process.platform}; ${process.arch})`;
   const baseHeaders = {
     "User-Agent": userAgent2
@@ -164241,9 +164379,36 @@ function getCoreSystemPrompt(userMemory, model) {
     }
   }
   const isGemini3 = model?.toLowerCase().includes("gemini-3-pro-preview") || model?.toLowerCase().includes("gemini-3.1-pro-preview") || model?.toLowerCase().includes("gemini-3-flash-preview");
+  const isGpt53Codex = model?.toLowerCase().includes("gpt-5.3-codex");
   const mandatesVariant = isGemini3 ? `
 
 - **Do not call tools in silence:** You must provide to the user very short and concise natural explanation (one sentence) before calling tools.` : ``;
+  const codexFinalAnswerInstructionsVariant = isGpt53Codex ? `
+
+## Final answer instructions
+- Balance conciseness to not overwhelm the user with appropriate detail for the request. Do not narrate abstractly; explain what you are doing and why.
+- Do not begin responses with conversational interjections or meta commentary. Avoid openers such as acknowledgements (\u201CDone \u2014\u201D, \u201CGot it\u201D, \u201CGreat question, \u201D) or framing phrases.
+- The user does not see command execution outputs. When asked to show the output of a command (e.g. \`git show\`), relay the important details in your answer or summarize the key lines so the user understands the result.
+- Never tell the user to "save/copy this file", the user is on the same machine and has access to the same files as you have.
+- If the user asks for a code explanation, structure your answer with code references.
+- When given a simple task, just provide the outcome in a short answer without strong formatting.
+- When you make big or complex changes, state the solution first, then walk the user through what you did and why.
+- For casual chit-chat, just chat.
+- If you weren't able to do something, for example run tests, tell the user.
+- If there are natural next steps the user may want to take, suggest them at the end of your response. Do not make suggestions if there are no natural next steps. When suggesting multiple options, use numeric lists for the suggestions so the user can quickly respond with a single number.` : ``;
+  const codexIntermediaryUpdatesVariant = isGpt53Codex ? `
+
+## Intermediary updates
+- User updates are short updates while you are working, they are NOT final answers.
+- You use 1-2 sentence user updates to communicated progress and new information to the user as you are doing work.
+- Do not begin responses with conversational interjections or meta commentary. Avoid openers such as acknowledgements (\u201CDone \u2014\u201D, \u201CGot it\u201D, \u201CGreat question, \u201D) or framing phrases.
+- You provide user updates frequently, every 20s.
+- Before exploring or doing substantial work, you start with a user update acknowledging the request and explaining your first step. You should include your understanding of the user request and explain what you will do. Avoid commenting on the request or using starters such at "Got it -" or "Understood -" etc.
+- When exploring, e.g. searching, reading files you provide user updates as you go, every 20s, explaining what context you are gathering and what you've learned. Vary your sentence structure when providing these updates to avoid sounding repetitive - in particular, don't start each sentence the same way.
+- After you have sufficient context, and the work is substantial you provide a longer plan (this is the only user update that may be longer than 2 sentences and can contain formatting).
+- Before performing file edits of any kind, you provide updates explaining what edits you are making.
+- As you are thinking, you very frequently provide updates even if not taking any actions, informing the user of your progress. You interrupt your thinking and send multiple updates in a row if thinking for more than 100 words.
+- Tone of your updates MUST match your personality.` : ``;
   const basePrompt = systemMdEnabled ? fs23.readFileSync(systemMdPath, "utf8") : `
 You are RDMind, an interactive CLI agent, specializing in software engineering tasks. Your primary goal is to help users safely and efficiently, adhering strictly to the following instructions and utilizing your available tools.
 
@@ -164379,6 +164544,9 @@ IMPORTANT: Always use the ${ToolNames.TODO_WRITE} tool to plan and track tasks t
 - **Help Command:** The user can use '/help' to display help information.
 - **Feedback:** To report a bug or provide feedback, please use the /bug command.
 
+${codexFinalAnswerInstructionsVariant}
+${codexIntermediaryUpdatesVariant}
+
 ${function() {
     const isSandboxExec = process9.env["SANDBOX"] === "sandbox-exec";
     const isGenericSandbox = !!process9.env["SANDBOX"];
@@ -251093,7 +251261,7 @@ function resolveModelConfig(input) {
     settings?.generationConfig,
     modelProvider?.generationConfig,
     authType,
-    modelProvider?.id,
+    modelProvider?.id ?? modelResult.value,
     sources
   );
   const config2 = {
@@ -251212,6 +251380,10 @@ function resolveGenerationConfig(settingsConfig, modelProviderConfig, authType,
       sources[field] = settingsSource(`model.generationConfig.${field}`);
     }
   }
+  if (result.modalities === void 0 && modelId) {
+    result.modalities = defaultModalities(modelId);
+    sources["modalities"] = computedSource("auto-detected from model");
+  }
   return result;
 }
 var init_modelConfigResolver = __esm({
@@ -251223,6 +251395,7 @@ var init_modelConfigResolver = __esm({
     init_apiKeyEncryption();
     init_configResolver();
     init_constants3();
+    init_modalityDefaults();
     init_contentGenerator();
     __name(resolveModelConfig, "resolveModelConfig");
     __name(resolveQwenOAuthConfig, "resolveQwenOAuthConfig");
@@ -361620,7 +361793,7 @@ __name(getPackageJson, "getPackageJson");
 // packages/cli/src/utils/version.ts
 async function getCliVersion() {
   const pkgJson = await getPackageJson();
-  return "0.2.8-alpha.0";
+  return "0.2.8-alpha.2";
 }
 __name(getCliVersion, "getCliVersion");
 
@@ -369369,7 +369542,7 @@ var formatDuration = /* @__PURE__ */ __name((milliseconds) => {
 
 // packages/cli/src/generated/git-commit.ts
 init_esbuild_shims();
-var GIT_COMMIT_INFO = "76368f06e";
+var GIT_COMMIT_INFO = "e33c3c8f2";
 
 // packages/cli/src/utils/systemInfo.ts
 async function getNpmVersion() {
@@ -402424,7 +402597,7 @@ function useTextBuffer({
   const openInExternalEditor = (0, import_react46.useCallback)(
     async (opts = {}) => {
       const editor = opts.editor ?? process.env["VISUAL"] ?? process.env["EDITOR"] ?? (process.platform === "win32" ? "notepad" : "vi");
-      const tmpDir = fs100.mkdtempSync(pathMod.join(os38.tmpdir(), "gemini-edit-"));
+      const tmpDir = fs100.mkdtempSync(pathMod.join(os38.tmpdir(), "rdmind-edit-"));
       const filePath = pathMod.join(tmpDir, "buffer.txt");
       fs100.writeFileSync(filePath, text, "utf8");
       dispatch({ type: "create_undo_snapshot" });
@@ -407680,7 +407853,7 @@ function getAuthDisplayType(authType, baseUrl, apiKeyEnvKey) {
   }
   switch (authType) {
     case "xhs-sso" /* XHS_SSO */:
-      return "XHS SSO" /* XHS_SSO */;
+      return "xhs-sso" /* XHS_SSO */;
     case "qwen-oauth" /* QWEN_OAUTH */:
       return "Qwen OAuth" /* QWEN_OAUTH */;
     default:
@@ -429288,7 +429461,7 @@ var GeminiAgent = class {
       name: APPROVAL_MODE_INFO[mode].name,
       description: APPROVAL_MODE_INFO[mode].description
     }));
-    const version2 = "0.2.8-alpha.0";
+    const version2 = "0.2.8-alpha.2";
     return {
       protocolVersion: PROTOCOL_VERSION,
       agentInfo: {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@rdmind/rdmind",
-  "version": "0.2.8-alpha.0",
+  "version": "0.2.8-alpha.2",
   "description": "RDMind - AI-powered coding assistant",
   "type": "module",
   "main": "cli.js",
@@ -19,7 +19,7 @@
     "locales"
   ],
   "config": {
-    "sandboxImageUri": "ghcr.io/qwenlm/qwen-code:0.2.8-alpha.0"
+    "sandboxImageUri": "ghcr.io/qwenlm/qwen-code:0.2.8-alpha.2"
   },
   "publishConfig": {
     "access": "public"