feat(proxy): enhance Anthropic-to-OpenAI message transformation and schema sanitization

src/proxy/transformers/request-transformer.ts

@@ -1,5 +1,7 @@
+import { normalizeSchemaForOpenAI } from '../../utils/schema-sanitizer';
+
 interface AnthropicThinking {
-  type?: 'enabled' | 'disabled' | string;
+  type?: 'enabled' | 'disabled' | 'adaptive' | string;
   budget_tokens?: number;
 }
 
@@ -14,6 +16,7 @@ interface AnthropicImageBlock {
     type?: string;
     media_type?: string;
     data?: string;
+    url?: string;
   };
 }
 
@@ -28,6 +31,7 @@ interface AnthropicToolResultBlock {
   type: 'tool_result';
   tool_use_id?: string;
   content?: unknown;
+  is_error?: boolean;
 }
 
 type AnthropicContentBlock =
@@ -42,6 +46,10 @@ interface AnthropicMessage {
   content?: string | AnthropicContentBlock[];
 }
 
+interface AnthropicOutputConfig {
+  effort?: 'low' | 'medium' | 'high' | 'max' | string;
+}
+
 interface AnthropicProxyRequestShape {
   model?: unknown;
   system?: unknown;
@@ -54,6 +62,7 @@ interface AnthropicProxyRequestShape {
   tools?: unknown;
   stream?: unknown;
   thinking?: AnthropicThinking;
+  output_config?: AnthropicOutputConfig;
 }
 
 interface OpenAITextPart {
@@ -108,7 +117,6 @@ export interface ProxyOpenAIRequest {
   metadata?: Record<string, unknown>;
 }
 
-const TOOL_RESULT_SERIALIZATION_FALLBACK = '[unserializable content]';
 const TOOL_USE_ARGUMENTS_FALLBACK = '{}';
 
 function assertObject(value: unknown, label: string): Record<string, unknown> {
@@ -168,17 +176,63 @@ function flattenTextContent(content: unknown, label: string): string {
     .join('\n');
 }
 
-function toToolResultContent(content: unknown, label: string): string {
+/**
+ * Convert tool_result content to OpenAI-compatible format.
+ * Handles strings, arrays with text/image blocks, and error prefixing.
+ * Ported from openclaude's convertToolResultContent.
+ */
+function convertToolResultContent(
+  content: unknown,
+  isError: boolean
+): string | OpenAIContentPart[] {
   if (content === undefined) {
     return '';
   }
   if (typeof content === 'string') {
-    return content;
+    return isError ? `Error: ${content}` : content;
   }
-  if (Array.isArray(content)) {
-    return flattenTextContent(content, label);
+  if (!Array.isArray(content)) {
+    const text = safeJsonStringify(content, '[unserializable content]');
+    return isError ? `Error: ${text}` : text;
   }
-  return safeJsonStringify(content, TOOL_RESULT_SERIALIZATION_FALLBACK);
+
+  const parts: OpenAIContentPart[] = [];
+  for (const block of content) {
+    if (block?.type === 'text' && typeof block.text === 'string') {
+      parts.push({ type: 'text', text: block.text });
+      continue;
+    }
+
+    if (block?.type === 'image') {
+      const source = block.source;
+      if (source?.type === 'url' && source.url) {
+        parts.push({ type: 'image_url', image_url: { url: source.url } });
+      } else if (source?.type === 'base64' && source.media_type && source.data) {
+        parts.push({
+          type: 'image_url',
+          image_url: { url: `data:${source.media_type};base64,${source.data}` },
+        });
+      }
+      continue;
+    }
+
+    if (typeof block?.text === 'string') {
+      parts.push({ type: 'text', text: block.text });
+    }
+  }
+
+  if (parts.length === 0) return '';
+  if (parts.length === 1 && parts[0].type === 'text') {
+    const text = (parts[0] as OpenAITextPart).text;
+    return isError ? `Error: ${text}` : text;
+  }
+  if (isError && parts[0]?.type === 'text') {
+    parts[0] = { ...parts[0], text: `Error: ${(parts[0] as OpenAITextPart).text}` };
+  } else if (isError) {
+    parts.unshift({ type: 'text', text: 'Error:' });
+  }
+
+  return parts;
 }
 
 function createFallbackToolId(messageIndex: number, blockIndex: number): string {
@@ -187,16 +241,27 @@ function createFallbackToolId(messageIndex: number, blockIndex: number): string
 
 function toImagePart(block: AnthropicImageBlock, label: string): OpenAIImagePart {
   const source = block.source;
-  if (!source || source.type !== 'base64' || !source.media_type || !source.data) {
-    throw new Error(`${label}.source must be a base64 image payload`);
+  if (!source) {
+    throw new Error(`${label}.source is missing`);
   }
 
-  return {
-    type: 'image_url',
-    image_url: {
-      url: `data:${source.media_type};base64,${source.data}`,
-    },
-  };
+  if (source.type === 'url' && source.url) {
+    return {
+      type: 'image_url',
+      image_url: { url: source.url },
+    };
+  }
+
+  if (source.type === 'base64' && source.media_type && source.data) {
+    return {
+      type: 'image_url',
+      image_url: {
+        url: `data:${source.media_type};base64,${source.data}`,
+      },
+    };
+  }
+
+  throw new Error(`${label}.source must be a base64 or url image payload`);
 }
 
 function isImageBlock(block: AnthropicContentBlock): block is AnthropicImageBlock {
@@ -234,30 +299,46 @@ function transformTools(value: unknown): ProxyOpenAIRequest['tools'] {
       (entry): entry is { name?: unknown; description?: unknown; input_schema?: unknown } =>
         typeof entry === 'object' && entry !== null
     )
-    .map((entry) => ({
-      type: 'function' as const,
-      function: {
-        name: typeof entry.name === 'string' ? entry.name : 'tool',
-        ...(typeof entry.description === 'string' ? { description: entry.description } : {}),
-        parameters:
-          typeof entry.input_schema === 'object' && entry.input_schema !== null
-            ? (entry.input_schema as Record<string, unknown>)
-            : { type: 'object', properties: {} },
-      },
-    }));
+    .map((entry) => {
+      const rawSchema =
+        typeof entry.input_schema === 'object' && entry.input_schema !== null
+          ? (entry.input_schema as Record<string, unknown>)
+          : { type: 'object', properties: {} };
+
+      return {
+        type: 'function' as const,
+        function: {
+          name: typeof entry.name === 'string' ? entry.name : 'tool',
+          ...(typeof entry.description === 'string' ? { description: entry.description } : {}),
+          parameters: normalizeSchemaForOpenAI(rawSchema),
+        },
+      };
+    });
 
   return tools.length > 0 ? tools : undefined;
 }
 
 function mapThinkingToReasoning(
-  thinking: AnthropicThinking | undefined
+  thinking: AnthropicThinking | undefined,
+  outputConfig: AnthropicOutputConfig | undefined
 ): Pick<ProxyOpenAIRequest, 'reasoning' | 'reasoning_effort'> {
   if (!thinking || thinking.type === 'disabled') {
     return {};
   }
 
+  if (thinking.type === 'adaptive') {
+    const effort = toOpenAIEffort(resolveOutputConfigEffort(outputConfig) ?? 'high');
+    return {
+      reasoning_effort: effort,
+      reasoning: {
+        enabled: true,
+        effort,
+      },
+    };
+  }
+
   if (thinking.type !== 'enabled') {
-    throw new Error('thinking.type must be "enabled" or "disabled"');
+    return {};
   }
 
   const effort =
@@ -274,6 +355,29 @@ function mapThinkingToReasoning(
   };
 }
 
+const VALID_EFFORT_LEVELS = new Set(['low', 'medium', 'high', 'max']);
+
+function resolveOutputConfigEffort(
+  outputConfig: AnthropicOutputConfig | undefined
+): string | undefined {
+  if (!outputConfig || typeof outputConfig.effort !== 'string') {
+    return undefined;
+  }
+  const normalized = outputConfig.effort.trim().toLowerCase();
+  return VALID_EFFORT_LEVELS.has(normalized) ? normalized : undefined;
+}
+
+/**
+ * Map Anthropic effort levels to OpenAI-compatible reasoning_effort.
+ * Anthropic's `max` has no standard OpenAI equivalent — most providers
+ * only accept low/medium/high and reject unknown values with a 400.
+ * Ported from openclaude's standardEffortToOpenAI() which maps max -> xhigh
+ * for Codex; for generic OpenAI-compat providers we clamp to high.
+ */
+function toOpenAIEffort(effort: string): string {
+  return effort === 'max' ? 'high' : effort;
+}
+
 function transformMessages(messagesValue: unknown): OpenAIMessage[] {
   if (!Array.isArray(messagesValue)) {
     throw new Error('messages must be an array');
@@ -298,10 +402,65 @@ function transformMessages(messagesValue: unknown): OpenAIMessage[] {
       throw new Error(`messages[${messageIndex}].content must be a string or array`);
     }
 
-    const userParts: OpenAIContentPart[] = [];
+    if (role === 'user') {
+      const userParts: OpenAIContentPart[] = [];
+      let sawToolResult = false;
+
+      content.forEach((block, blockIndex) => {
+        const parsed = assertObject(
+          block,
+          `messages[${messageIndex}].content[${blockIndex}]`
+        ) as AnthropicContentBlock;
+
+        if (parsed.type === 'thinking' || parsed.type === 'redacted_thinking') {
+          return;
+        }
+
+        if (parsed.type === 'text') {
+          const text = typeof parsed.text === 'string' ? parsed.text : '';
+          userParts.push({ type: 'text', text });
+          return;
+        }
+
+        if (isImageBlock(parsed)) {
+          userParts.push(toImagePart(parsed, `messages[${messageIndex}].content[${blockIndex}]`));
+          return;
+        }
+
+        if (isToolResultBlock(parsed)) {
+          if (typeof parsed.tool_use_id !== 'string' || parsed.tool_use_id.trim().length === 0) {
+            throw new Error(
+              `messages[${messageIndex}].content[${blockIndex}].tool_use_id must be a non-empty string`
+            );
+          }
+          sawToolResult = true;
+          flushUserContent(translatedMessages, userParts);
+          translatedMessages.push({
+            role: 'tool',
+            tool_call_id: parsed.tool_use_id,
+            content: convertToolResultContent(parsed.content, parsed.is_error === true),
+          });
+          return;
+        }
+
+        if (isToolUseBlock(parsed)) {
+          return;
+        }
+
+        throw new Error(
+          `messages[${messageIndex}].content[${blockIndex}].type "${String(parsed.type)}" is not supported`
+        );
+      });
+
+      if (userParts.length > 0 || !sawToolResult) {
+        flushUserContent(translatedMessages, userParts);
+      }
+      return;
+    }
+
+    // Assistant role
     const assistantTextParts: string[] = [];
     const toolCalls: NonNullable<OpenAIMessage['tool_calls']> = [];
-    let sawToolResult = false;
 
     content.forEach((block, blockIndex) => {
       const parsed = assertObject(
@@ -309,32 +468,17 @@ function transformMessages(messagesValue: unknown): OpenAIMessage[] {
         `messages[${messageIndex}].content[${blockIndex}]`
       ) as AnthropicContentBlock;
 
-      if (parsed.type === 'text') {
-        const text = typeof parsed.text === 'string' ? parsed.text : '';
-        if (role === 'user') {
-          userParts.push({ type: 'text', text });
-        } else {
-          assistantTextParts.push(text);
-        }
+      if (parsed.type === 'thinking' || parsed.type === 'redacted_thinking') {
         return;
       }
 
-      if (isImageBlock(parsed)) {
-        if (role !== 'user') {
-          throw new Error(
-            `messages[${messageIndex}].content[${blockIndex}] image requires user role`
-          );
-        }
-        userParts.push(toImagePart(parsed, `messages[${messageIndex}].content[${blockIndex}]`));
+      if (parsed.type === 'text') {
+        const text = typeof parsed.text === 'string' ? parsed.text : '';
+        assistantTextParts.push(text);
         return;
       }
 
       if (isToolUseBlock(parsed)) {
-        if (role !== 'assistant') {
-          throw new Error(
-            `messages[${messageIndex}].content[${blockIndex}] tool_use requires assistant role`
-          );
-        }
         toolCalls.push({
           id:
             typeof parsed.id === 'string' && parsed.id.length > 0
@@ -349,52 +493,61 @@ function transformMessages(messagesValue: unknown): OpenAIMessage[] {
         return;
       }
 
-      if (isToolResultBlock(parsed)) {
-        if (role !== 'user') {
-          throw new Error(
-            `messages[${messageIndex}].content[${blockIndex}] tool_result requires user role`
-          );
-        }
-        if (typeof parsed.tool_use_id !== 'string' || parsed.tool_use_id.trim().length === 0) {
-          throw new Error(
-            `messages[${messageIndex}].content[${blockIndex}].tool_use_id must be a non-empty string`
-          );
-        }
-        sawToolResult = true;
-        flushUserContent(translatedMessages, userParts);
-        translatedMessages.push({
-          role: 'tool',
-          tool_call_id: parsed.tool_use_id,
-          content: toToolResultContent(
-            parsed.content,
-            `messages[${messageIndex}].content[${blockIndex}].content`
-          ),
-        });
+      if (isImageBlock(parsed) || isToolResultBlock(parsed)) {
         return;
       }
-
-      throw new Error(
-        `messages[${messageIndex}].content[${blockIndex}].type "${String(parsed.type)}" is not supported`
-      );
     });
 
-    if (role === 'assistant') {
-      translatedMessages.push({
-        role: 'assistant',
-        content: assistantTextParts.join('\n'),
-        tool_calls: toolCalls.length > 0 ? toolCalls : undefined,
-      });
-      return;
-    }
-
-    if (userParts.length > 0 || !sawToolResult) {
-      flushUserContent(translatedMessages, userParts);
-    }
+    translatedMessages.push({
+      role: 'assistant',
+      content: assistantTextParts.join('\n'),
+      tool_calls: toolCalls.length > 0 ? toolCalls : undefined,
+    });
   });
 
   return translatedMessages;
 }
 
+/**
+ * Coalesce consecutive messages of the same role.
+ * OpenAI/vLLM/Ollama/Mistral require strict user<->assistant alternation.
+ * Multiple consecutive tool messages are allowed (assistant -> tool* -> user).
+ * Ported from openclaude's coalescing pass.
+ */
+function coalesceMessages(messages: OpenAIMessage[]): OpenAIMessage[] {
+  const coalesced: OpenAIMessage[] = [];
+
+  for (const msg of messages) {
+    const prev = coalesced[coalesced.length - 1];
+
+    if (prev && prev.role === msg.role && msg.role !== 'tool' && msg.role !== 'system') {
+      const prevContent = prev.content;
+      const curContent = msg.content;
+
+      if (typeof prevContent === 'string' && typeof curContent === 'string') {
+        prev.content = prevContent + (prevContent && curContent ? '\n' : '') + curContent;
+      } else {
+        const toArray = (
+          c: string | OpenAIContentPart[] | null | undefined
+        ): OpenAIContentPart[] => {
+          if (!c) return [];
+          if (typeof c === 'string') return c ? [{ type: 'text', text: c }] : [];
+          return c;
+        };
+        prev.content = [...toArray(prevContent), ...toArray(curContent)];
+      }
+
+      if (msg.tool_calls?.length) {
+        prev.tool_calls = [...(prev.tool_calls ?? []), ...msg.tool_calls];
+      }
+    } else {
+      coalesced.push({ ...msg });
+    }
+  }
+
+  return coalesced;
+}
+
 export class ProxyRequestTransformer {
   transform(raw: unknown): ProxyOpenAIRequest {
     const source = assertObject(raw || {}, 'request') as AnthropicProxyRequestShape;
@@ -414,14 +567,14 @@ export class ProxyRequestTransformer {
           ? source.model.trim()
           : undefined,
       stream: source.stream === true,
-      messages: allMessages,
+      messages: coalesceMessages(allMessages),
       max_tokens: asNumber(source.max_tokens),
       temperature: asNumber(source.temperature),
       top_p: asNumber(source.top_p),
       stop: asStringArray(source.stop_sequences),
       metadata: asMetadata(source.metadata),
       tools: transformTools(source.tools),
-      ...mapThinkingToReasoning(source.thinking),
+      ...mapThinkingToReasoning(source.thinking, source.output_config),
     };
   }
 }