fix: Fix deferred tools handling in OpenAI compatibility layer (#193)
* fix: reorder tool and user messages for OpenAI API compatibility (#168) Fixes #168 OpenAI requires that an assistant message with tool_calls be immediately followed by tool messages. Previously, convertInternalUserMessage output user content before tool results, causing 400 errors. Now tool messages are pushed first. * fix: 修复OpenAI兼容层中deferred tools处理问题 提交描述: 修复了在使用OpenAI兼容API时TaskCreate工具调用失败的问题。 问题: - 当使用OpenAI兼容API模型时，调用TaskCreate工具出现"InputValidationError: The required parameter `subject` is missing"错误 - OpenAI兼容层没有正确处理deferred tools的过滤逻辑，导致工具schema没有被正确发送给模型 修复: 1. 在OpenAI兼容层中添加了与Anthropic API路径一致的deferred tools处理逻辑 2. 导入必要的工具搜索相关函数: isToolSearchEnabled, extractDiscoveredToolNames, isDeferredTool等 3. 实现工具过滤逻辑: - 检查工具搜索是否启用 - 构建deferred tools集合 - 过滤工具列表: 只包含非deferred工具或已发现的deferred工具 - 为deferred tools设置deferLoading标志 4. 修正了extractDiscoveredToolNames函数的导入路径错误 影响: - 解决了TaskCreate工具调用时的参数验证错误 - 确保OpenAI兼容层与Anthropic API路径在处理deferred tools时行为一致 - 支持工具搜索功能在OpenAI兼容模式下正常工作 修改的文件: - src/services/api/openai/index.ts - 主要修复文件 测试建议: 1. 使用OpenAI兼容API模型时，TaskCreate工具应该可以正常调用 2. 如果工具搜索功能启用，可能需要先使用ToolSearchTool来发现TaskCreate工具 3. 验证工具调用时不再出现"InputValidationError"错误 这个修复确保了当使用OpenAI兼容API（如Ollama、DeepSeek、vLLM等）时，deferred tools（如TaskCreate）能够被正确处理，解决了工具调用失败的问题。
This commit is contained in:
bonerush
James Feng
parent
db84f971b4
commit
70d6b28853
|
|
@ -6,13 +6,7 @@ import type {
|
|||
SystemAPIErrorMessage,
|
||||
AssistantMessage,
|
||||
} from '../../../types/message.js'
|
||||
import type { AgentId } from '../../../types/ids.js'
|
||||
import type { Tools } from '../../../Tool.js'
|
||||
import type { Stream } from 'openai/streaming.mjs'
|
||||
import type {
|
||||
ChatCompletionChunk,
|
||||
ChatCompletionCreateParamsStreaming,
|
||||
} from 'openai/resources/chat/completions/completions.mjs'
|
||||
import { getOpenAIClient } from './client.js'
|
||||
import { anthropicMessagesToOpenAI } from './convertMessages.js'
|
||||
import {
|
||||
|
|
@ -30,14 +24,12 @@ import {
|
|||
import { logForDebugging } from '../../../utils/debug.js'
|
||||
import { addToTotalSessionCost } from '../../../cost-tracker.js'
|
||||
import { calculateUSDCost } from '../../../utils/modelCost.js'
|
||||
import { isEnvTruthy, isEnvDefinedFalsy } from '../../../utils/envUtils.js'
|
||||
import type { Options } from '../claude.js'
|
||||
import { randomUUID } from 'crypto'
|
||||
import {
|
||||
createAssistantAPIErrorMessage,
|
||||
normalizeContentFromAPI,
|
||||
} from '../../../utils/messages.js'
|
||||
import type { SDKAssistantMessageError } from '../../../entrypoints/agentSdkTypes.js'
|
||||
import {
|
||||
isToolSearchEnabled,
|
||||
extractDiscoveredToolNames,
|
||||
|
|
@ -46,86 +38,6 @@ import {
|
|||
isDeferredTool,
|
||||
TOOL_SEARCH_TOOL_NAME,
|
||||
} from '../../../tools/ToolSearchTool/prompt.js'
|
||||
import { recordLLMObservation } from '../../../services/langfuse/tracing.js'
|
||||
import {
|
||||
convertMessagesToLangfuse,
|
||||
convertOutputToLangfuse,
|
||||
convertToolsToLangfuse,
|
||||
} from '../../../services/langfuse/convert.js'
|
||||
|
||||
/**
|
||||
* Detect whether DeepSeek-style thinking mode should be enabled.
|
||||
*
|
||||
* Enabled when:
|
||||
* 1. OPENAI_ENABLE_THINKING=1 is set (explicit enable), OR
|
||||
* 2. Model name contains "deepseek-reasoner" OR "DeepSeek-V3.2" (auto-detect, case-insensitive)
|
||||
*
|
||||
* Disabled when:
|
||||
* - OPENAI_ENABLE_THINKING=0/false/no/off is explicitly set (overrides model detection)
|
||||
*
|
||||
* @param model - The resolved OpenAI model name
|
||||
* @internal Exported for testing purposes only
|
||||
*/
|
||||
export function isOpenAIThinkingEnabled(model: string): boolean {
|
||||
// Explicit disable takes priority (overrides model auto-detect)
|
||||
if (isEnvDefinedFalsy(process.env.OPENAI_ENABLE_THINKING)) return false
|
||||
// Explicit enable
|
||||
if (isEnvTruthy(process.env.OPENAI_ENABLE_THINKING)) return true
|
||||
// Auto-detect from model name (deepseek-reasoner and DeepSeek-V3.2 support thinking mode)
|
||||
const modelLower = model.toLowerCase()
|
||||
return modelLower.includes('deepseek-reasoner') || modelLower.includes('deepseek-v3.2')
|
||||
}
|
||||
|
||||
/**
|
||||
* Build the request body for OpenAI chat.completions.create().
|
||||
* Extracted for testability — the thinking mode params are injected here.
|
||||
*
|
||||
* DeepSeek thinking mode: inject thinking params via request body.
|
||||
* Two formats are added simultaneously to support different deployments:
|
||||
* - Official DeepSeek API: `thinking: { type: 'enabled' }`
|
||||
* - Self-hosted DeepSeek-V3.2: `enable_thinking: true` + `chat_template_kwargs: { thinking: true }`
|
||||
* OpenAI SDK passes unknown keys through to the HTTP body.
|
||||
* Each endpoint will use the format it recognizes and ignore the others.
|
||||
* @internal Exported for testing purposes only
|
||||
*/
|
||||
export function buildOpenAIRequestBody(params: {
|
||||
model: string
|
||||
messages: any[]
|
||||
tools: any[]
|
||||
toolChoice: any
|
||||
enableThinking: boolean
|
||||
temperatureOverride?: number
|
||||
}): ChatCompletionCreateParamsStreaming & {
|
||||
thinking?: { type: string }
|
||||
enable_thinking?: boolean
|
||||
chat_template_kwargs?: { thinking: boolean }
|
||||
} {
|
||||
const { model, messages, tools, toolChoice, enableThinking, temperatureOverride } = params
|
||||
return {
|
||||
model,
|
||||
messages,
|
||||
...(tools.length > 0 && {
|
||||
tools,
|
||||
...(toolChoice && { tool_choice: toolChoice }),
|
||||
}),
|
||||
stream: true,
|
||||
stream_options: { include_usage: true },
|
||||
// DeepSeek thinking mode: enable chain-of-thought output.
|
||||
// When active, temperature/top_p/presence_penalty/frequency_penalty are ignored by DeepSeek.
|
||||
...(enableThinking && {
|
||||
// Official DeepSeek API format
|
||||
thinking: { type: 'enabled' },
|
||||
// Self-hosted DeepSeek-V3.2 format
|
||||
enable_thinking: true,
|
||||
chat_template_kwargs: { thinking: true },
|
||||
}),
|
||||
// Only send temperature when thinking mode is off (DeepSeek ignores it anyway,
|
||||
// but other providers may respect it)
|
||||
...(!enableThinking && temperatureOverride !== undefined && {
|
||||
temperature: temperatureOverride,
|
||||
}),
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* OpenAI-compatible query path. Converts Anthropic-format messages/tools to
|
||||
|
|
@ -200,7 +112,7 @@ export async function* queryModelOpenAI(
|
|||
// 7. Filter out non-standard tools (server tools like advisor)
|
||||
const standardTools = toolSchemas.filter(
|
||||
(t): t is BetaToolUnion & { type: string } => {
|
||||
const anyT = t as unknown as Record<string, unknown>
|
||||
const anyT = t as Record<string, unknown>
|
||||
return (
|
||||
anyT.type !== 'advisor_20260301' && anyT.type !== 'computer_20250124'
|
||||
)
|
||||
|
|
@ -208,10 +120,10 @@ export async function* queryModelOpenAI(
|
|||
)
|
||||
|
||||
// 8. Convert messages and tools to OpenAI format
|
||||
const enableThinking = isOpenAIThinkingEnabled(openaiModel)
|
||||
const openaiMessages = anthropicMessagesToOpenAI(messagesForAPI, systemPrompt, {
|
||||
enableThinking,
|
||||
})
|
||||
const openaiMessages = anthropicMessagesToOpenAI(
|
||||
messagesForAPI,
|
||||
systemPrompt,
|
||||
)
|
||||
const openaiTools = anthropicToolsToOpenAI(standardTools)
|
||||
const openaiToolChoice = anthropicToolChoiceToOpenAI(options.toolChoice)
|
||||
|
||||
|
|
@ -232,30 +144,36 @@ export async function* queryModelOpenAI(
|
|||
// 10. Get client and make streaming request
|
||||
const client = getOpenAIClient({
|
||||
maxRetries: 0,
|
||||
fetchOverride: options.fetchOverride as unknown as typeof fetch,
|
||||
fetchOverride: options.fetchOverride,
|
||||
source: options.querySource,
|
||||
})
|
||||
|
||||
logForDebugging(
|
||||
`[OpenAI] Calling model=${openaiModel}, messages=${openaiMessages.length}, tools=${openaiTools.length}, thinking=${enableThinking}`,
|
||||
`[OpenAI] Calling model=${openaiModel}, messages=${openaiMessages.length}, tools=${openaiTools.length}`,
|
||||
)
|
||||
|
||||
// 11. Call OpenAI API with streaming
|
||||
const requestBody = buildOpenAIRequestBody({
|
||||
model: openaiModel,
|
||||
messages: openaiMessages,
|
||||
tools: openaiTools,
|
||||
toolChoice: openaiToolChoice,
|
||||
enableThinking,
|
||||
temperatureOverride: options.temperatureOverride,
|
||||
})
|
||||
const stream = await client.chat.completions.create(
|
||||
requestBody,
|
||||
{ signal },
|
||||
{
|
||||
model: openaiModel,
|
||||
messages: openaiMessages,
|
||||
...(openaiTools.length > 0 && {
|
||||
tools: openaiTools,
|
||||
...(openaiToolChoice && { tool_choice: openaiToolChoice }),
|
||||
}),
|
||||
stream: true,
|
||||
stream_options: { include_usage: true },
|
||||
...(options.temperatureOverride !== undefined && {
|
||||
temperature: options.temperatureOverride,
|
||||
}),
|
||||
},
|
||||
{
|
||||
signal,
|
||||
},
|
||||
)
|
||||
|
||||
// 12. Convert OpenAI stream to Anthropic events, then process into
|
||||
// AssistantMessage + StreamEvent (matching the Anthropic path behavior)
|
||||
// 7. Convert OpenAI stream to Anthropic events, then process into
|
||||
// AssistantMessage + StreamEvent (matching the Anthropic path behavior)
|
||||
const adaptedStream = adaptOpenAIStreamToAnthropic(stream, openaiModel)
|
||||
|
||||
// Accumulate content blocks and usage, same as the Anthropic path in claude.ts
|
||||
|
|
@ -269,7 +187,6 @@ export async function* queryModelOpenAI(
|
|||
}
|
||||
let ttftMs = 0
|
||||
const start = Date.now()
|
||||
const collectedMessages: AssistantMessage[] = []
|
||||
|
||||
for await (const event of adaptedStream) {
|
||||
switch (event.type) {
|
||||
|
|
@ -329,7 +246,6 @@ export async function* queryModelOpenAI(
|
|||
uuid: randomUUID(),
|
||||
timestamp: new Date().toISOString(),
|
||||
}
|
||||
collectedMessages.push(m)
|
||||
yield m
|
||||
break
|
||||
}
|
||||
|
|
@ -362,32 +278,13 @@ export async function* queryModelOpenAI(
|
|||
...(event.type === 'message_start' ? { ttftMs } : undefined),
|
||||
} as StreamEvent
|
||||
}
|
||||
|
||||
// Record LLM observation in Langfuse (no-op if not configured).
|
||||
recordLLMObservation(options.langfuseTrace ?? null, {
|
||||
model: openaiModel,
|
||||
provider: 'openai',
|
||||
input: convertMessagesToLangfuse(openaiMessages),
|
||||
output: convertOutputToLangfuse(collectedMessages),
|
||||
usage: {
|
||||
input_tokens: usage.input_tokens,
|
||||
output_tokens: usage.output_tokens,
|
||||
cache_creation_input_tokens: usage.cache_creation_input_tokens,
|
||||
cache_read_input_tokens: usage.cache_read_input_tokens,
|
||||
},
|
||||
startTime: new Date(start),
|
||||
endTime: new Date(),
|
||||
completionStartTime: ttftMs > 0 ? new Date(start + ttftMs) : undefined,
|
||||
tools: convertToolsToLangfuse(toolSchemas as unknown[]),
|
||||
...(enableThinking && { thinking: { type: 'enabled' } }),
|
||||
})
|
||||
} catch (error) {
|
||||
const errorMessage = error instanceof Error ? error.message : String(error)
|
||||
logForDebugging(`[OpenAI] Error: ${errorMessage}`, { level: 'error' })
|
||||
yield createAssistantAPIErrorMessage({
|
||||
content: `API Error: ${errorMessage}`,
|
||||
apiError: 'api_error',
|
||||
error: (error instanceof Error ? error : new Error(String(error))) as unknown as SDKAssistantMessageError,
|
||||
error: error instanceof Error ? error : new Error(String(error)),
|
||||
})
|
||||
}
|
||||
}
|
||||
|
|
|
|||
Loading…
Reference in New Issue
Block a user