LLM/CLIProxyAPIPlus
1
0
Fork 0
mirror of https://github.com/router-for-me/CLIProxyAPIPlus.git synced 2026-04-25 08:36:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(kiro): implement official reasoningContentEvent and improve metadat

Browse Source
This commit is contained in:
Ravens2121
2025-12-18 04:38:22 +08:00
parent 0155a01bb1
commit d687ee2777
3 changed files with 431 additions and 477 deletions
Download Patch File Download Diff File Expand all files Collapse all files

878
internal/runtime/executor/kiro_executor.go
View File

File diff suppressed because it is too large Load Diff

15
internal/translator/kiro/claude/kiro_claude_request.go
View File

@@ -222,20 +222,19 @@ func BuildKiroPayload(claudeBody []byte, modelID, profileArn, origin string, isA
kiroTools := convertClaudeToolsToKiro(tools)
// Thinking mode implementation:
// Kiro API doesn't accept max_tokens for thinking. Instead, thinking mode is enabled
// by injecting <thinking_mode> and <max_thinking_length> tags into the system prompt.
// We use a fixed max_thinking_length value since Kiro handles the actual budget internally.
// Kiro API supports official thinking/reasoning mode via <thinking_mode> tag.
// When set to "enabled", Kiro returns reasoning content as official reasoningContentEvent
// rather than inline <thinking> tags in assistantResponseEvent.
// We use a high max_thinking_length to allow extensive reasoning.
if thinkingEnabled {
thinkingHint := `<thinking_mode>interleaved</thinking_mode>
<max_thinking_length>200000</max_thinking_length>
IMPORTANT: You MUST use <thinking>...</thinking> tags to show your reasoning process before providing your final response. Think step by step inside the thinking tags.`
thinkingHint := `<thinking_mode>enabled</thinking_mode>
<max_thinking_length>200000</max_thinking_length>`
if systemPrompt != "" {
systemPrompt = thinkingHint + "\n\n" + systemPrompt
} else {
systemPrompt = thinkingHint
}
log.Infof("kiro: injected thinking prompt, has_tools: %v", len(kiroTools) > 0)
log.Infof("kiro: injected thinking prompt (official mode), has_tools: %v", len(kiroTools) > 0)
}
// Process messages and build history

15
internal/translator/kiro/openai/kiro_openai_request.go
View File

@@ -231,20 +231,19 @@ func BuildKiroPayloadFromOpenAI(openaiBody []byte, modelID, profileArn, origin s
kiroTools := convertOpenAIToolsToKiro(tools)
// Thinking mode implementation:
// Kiro API doesn't accept max_tokens for thinking. Instead, thinking mode is enabled
// by injecting <thinking_mode> and <max_thinking_length> tags into the system prompt.
// We use a fixed max_thinking_length value since Kiro handles the actual budget internally.
// Kiro API supports official thinking/reasoning mode via <thinking_mode> tag.
// When set to "enabled", Kiro returns reasoning content as official reasoningContentEvent
// rather than inline <thinking> tags in assistantResponseEvent.
// We use a high max_thinking_length to allow extensive reasoning.
if thinkingEnabled {
thinkingHint := `<thinking_mode>interleaved</thinking_mode>
<max_thinking_length>200000</max_thinking_length>
IMPORTANT: You MUST use <thinking>...</thinking> tags to show your reasoning process before providing your final response. Think step by step inside the thinking tags.`
thinkingHint := `<thinking_mode>enabled</thinking_mode>
<max_thinking_length>200000</max_thinking_length>`
if systemPrompt != "" {
systemPrompt = thinkingHint + "\n\n" + systemPrompt
} else {
systemPrompt = thinkingHint
}
log.Debugf("kiro-openai: injected thinking prompt")
log.Debugf("kiro-openai: injected thinking prompt (official mode)")
}
// Process messages and build history