GitLab

injectClaudeCodePrompt 和 systemIncludesClaudeCodePrompt 的 type switch
无法匹配 json.RawMessage 类型（Go typed nil 陷阱），导致 ForwardAsResponses
和 ForwardAsChatCompletions 路径中用户 system prompt 被替换为仅 Claude Code
banner。新增 normalizeSystemParam 将 json.RawMessage 转为标准 Go 类型。
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

新增功能：
- 新增 TLS 指纹 Profile CRUD 管理（Ent schema + 迁移 + Admin API + 前端管理界面）
- 支持账号绑定数据库中的自定义 TLS Profile，或随机选择（profile_id=-1）
- HTTPUpstream.DoWithTLS 接口从 bool 改为 *tlsfingerprint.Profile，支持按账号指定 Profile
- AccountUsageService 注入 TLSFingerprintProfileService，统一 usage 场景与网关的 Profile 解析逻辑

代码优化：
- 删除已被 TLSFingerprintProfileService 完全取代的 registry.go 死代码（418 行）
- 提取 3 个 dialer 的重复 TLS 握手逻辑为 performTLSHandshake() 共用函数
- 修复 GetTLSFingerprintProfileID 缺少 json.Number 处理的 bug
- gateway_service.Forward 中 ResolveTLSProfile 从重试循环内重复调用改为预解析局部变量
- 删除冗余的 buildClientHelloSpec() 单行 wrapper 和 int64(e.ID) 无效转换
- tls_fingerprint_profile_cache.go 日志从 log.Printf 改为 slog 结构化日志
- dialer_capture_test.go 添加 //go:build integration 标签，防止 CI 失败
- 去重 TestProfileExpectation 类型至共享 test_types_test.go
- 修复 9 个测试文件缺少 tlsfingerprint import 的编译错误
- 修复 error_policy_integration_test.go 中 handleError 回调签名被错误替换的问题

新增独立开关控制 API Key 账号的签名整流功能，支持配置自定义
匹配关键词以捕获不同格式的上游错误响应。

- 新增 apikey_signature_enabled 开关（默认关闭）
- 新增 apikey_signature_patterns 自定义关键词配置
- 内置签名检测规则对 API Key 账号同样生效
- 自定义关键词对完整响应体做不区分大小写匹配
- 重试二阶段检测仅做模式匹配，不重复校验开关
- Handler 层校验关键词数量（≤50）和长度（≤500）
- API 响应 nil patterns 统一序列化为空数组
- OAuth/SetupToken/Upstream/Bedrock 账号行为不变

- 新增 header_util.go，通过 setHeaderRaw/getHeaderRaw/addHeaderRaw 绕过
  Go 的 canonical-case 规范化，保持真实 Claude CLI 抓包的请求头大小写
  （如 "x-app" 而非 "X-App"，"X-Stainless-OS" 而非 "X-Stainless-Os"）
- 新增管理后台开关：指纹统一化（默认开启）和 metadata 透传（默认关闭），
  使用 atomic.Value + singleflight 缓存模式，60s TTL
- 调试日志从控制台 body 打印升级为文件级完整快照
  （按真实 wire 顺序输出 headers + 格式化 JSON body + 上下文元数据）
- 恢复 accept-encoding 到白名单，在 http_upstream.go 新增 decompressResponseBody
  处理 gzip/brotli/deflate 解压（Go 显式设置 Accept-Encoding 时不会自动解压）
- OAuth 服务 axios UA 从 1.8.4 更新至 1.13.6
- 测试断言改用 getHeaderRaw 适配 raw header 存储方式

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent

)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

Empty text blocks inside tool_result.content were not being filtered,
causing upstream 400 errors: 'text content blocks must be non-empty'.

Changes:
- Add stripEmptyTextBlocksFromSlice helper for recursive content filtering
- FilterThinkingBlocksForRetry now recurses into tool_result nested content
- Add StripEmptyTextBlocks pre-filter on initial request path to avoid
  unnecessary 400+retry round-trips
- Add unit tests for nested empty text block scenarios

feat(ops): propagate endpoint/request-type context in handlers; add UpstreamURL to upstream error events

LegacyPrintf uses inferStdLogLevel() to infer log level from message
text. Any message containing the word "error" is classified as ERROR
level, causing the entire signature-retry recovery flow (which succeeds)
to produce spurious ERROR log entries.

Changes:
- Remove noisy [SignatureCheck] debug logs inside isThinkingBlockSignatureError
  that were logging every detected signature check as ERROR
- Change retry-start log to WARN level via [warn] prefix
- Change retry-success log to INFO level by removing "error" from message
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent

)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

Empty text blocks ({"type":"text","text":""}) cause Anthropic upstream to
return 400: "text content blocks must be non-empty". This was not caught
by the existing error detection pattern in isThinkingBlockSignatureError,
nor handled by FilterThinkingBlocksForRetry.

- Add empty text block stripping to FilterThinkingBlocksForRetry
- Fix isThinkingBlockSignatureError to match new Anthropic error format
- Add fast-path byte patterns to avoid unnecessary JSON parsing
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Claude Code v2.1.78 起将 metadata.user_id 从拼接字符串改为 JSON：
旧: user_{hex}_account_{uuid}_session_{uuid}
新: {"device_id":"...","account_uuid":"...","session_id":"..."}

新增集中解析/格式化模块 metadata_userid.go：
- ParseMetadataUserID: 自动识别两种格式，提取 DeviceID/AccountUUID/SessionID
- FormatMetadataUserID: 根据 UA 版本输出对应格式（>= 2.1.78 输出 JSON）
- ExtractCLIVersion: 从 UA 提取版本号，消除与 ClaudeCodeValidator.ExtractVersion 的重复

修改消费者统一使用新模块：
- claude_code_validator: 用 ParseMetadataUserID 替代只匹配旧格式的 userIDPattern
- identity_service: RewriteUserID/WithMasking 增加 fingerprintUA 参数，
  解析用 ParseMetadataUserID，输出用 FormatMetadataUserID（版本感知）
- gateway_service: GenerateSessionHash 用 ParseMetadataUserID 提取 session_id，
  buildOAuthMetadataUserID 用 FormatMetadataUserID 输出版本匹配格式，
  两处 RewriteUserIDWithMasking 调用传入 fp.UserAgent
- account_test_service: generateSessionString 改用 FormatMetadataUserID，
  自动跟随 DefaultHeaders UA 版本

删除三个旧正则: userIDPattern, userIDRegex, sessionIDRegex
统一 hex 匹配为 [a-fA-F0-9]，修复旧 userIDRegex 只匹配小写的不一致

Propagate UpstreamModel through ForwardResult and OpenAIForwardResult in Anthropic direct, API-key passthrough, Bedrock, and OpenAI gateway flows. Extract optionalNonEqualStringPtr and optionalTrimmedStringPtr into usage_log_helpers.go. Store upstream_model only when it differs from the requested model.

Also introduces anthropicPassthroughForwardInput struct to reduce parameter count.

Extend RecordUsageInput and RecordUsageLongContextInput structs with InboundEndpoint and UpstreamEndpoint so that Claude, Gemini, and Sora handlers can record endpoint info alongside OpenAI handlers.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode

)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

- Add 5th error filter switch IgnoreInsufficientBalanceErrors to suppress
  upstream insufficient balance / insufficient_quota errors from ops log
- Extract hardcoded error strings into package-level constants for
  shouldSkipOpsErrorLog, normalizeOpsErrorType, classifyOpsPhase, and
  classifyOpsIsBusinessLimited
- Define ErrNoAvailableAccounts sentinel error and replace all
  errors.New("no available accounts") call sites
- Update tests to use require.ErrorIs with the sentinel error

Claude's output_config.effort parameter (low/medium/high/max) was not
being extracted from requests or logged in the reasoning_effort column
of usage logs. Only the OpenAI path populated this field.

Changes:
- Extract output_config.effort in ParseGatewayRequest
- Add ReasoningEffort field to ForwardResult
- Populate reasoning_effort in both RecordUsage and RecordUsageWithLongContext
- Guard against overwriting service-set effort values in handler
- Update stale comments that described reasoning_effort as OpenAI-only
- Add unit tests for extraction, normalization, and persistence

Consolidate two separate channel types (bedrock + bedrock-apikey) into
a single "AWS Bedrock" channel. Authentication mode is now distinguished
by credentials.auth_mode ("sigv4" | "apikey") instead of separate types.

Backend:
- Remove AccountTypeBedrockAPIKey constant
- IsBedrock() simplified; IsBedrockAPIKey() checks auth_mode
- Add IsAPIKeyOrBedrock() helper to eliminate repeated type checks
- Extend pool mode, quota scheduling, and billing to bedrock
- Add RetryableOnSameAccount to handleBedrockUpstreamErrors
- Add "bedrock" scope to Beta Policy for independent control

Frontend:
- Merge two buttons into one "AWS Bedrock" with auth mode radio
- Badge displays "Anthropic | AWS"
- Pool mode and quota limit UI available for bedrock
- Quota display in account list (usage bars, capacity badges, reset)
- Remove all bedrock-apikey type references

Anthropic Messages API 的流式转发路径（gateway_service.go）在上游长时间
无数据时（如 Opus extended thinking 阶段）不会向下游发送任何内容，导致
Cloudflare Tunnel 等代理因连接空闲而断开。

复用已有的 StreamKeepaliveInterval 配置（默认 10 秒），在 select 循环中
添加 keepalive 分支，定时发送 Anthropic 原生格式的 ping 事件保活，与
OpenAI 兼容路径的实现模式保持一致。
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

4K image base64 data can exceed 40MB limit, causing "bufio.Scanner:
token too long" errors. Scanner is adaptive (starts at 64KB, grows
as needed), so increasing the cap has no impact on normal responses.

Extend the existing total quota limit with daily and weekly periodic
dimensions. Each dimension is independently configurable and uses lazy
reset — when the period expires, usage is automatically reset to zero on
the next increment. Any dimension exceeding its limit will pause the
account from scheduling.

Backend:
- Add GetQuotaDailyLimit/Used, GetQuotaWeeklyLimit/Used, HasAnyQuotaLimit
- Rewrite IncrementQuotaUsed with atomic CTE SQL for 3-dimension update
- Rewrite ResetQuotaUsed to clear all dimensions and period timestamps
- Update postUsageBilling to use HasAnyQuotaLimit()
- Preserve daily/weekly used values on account edit

Frontend:
- Refactor QuotaLimitCard from single v-model to 3-dimension props
- Add QuotaBadge component for compact D/W/$ display
- Update AccountCapacityCell with per-dimension badges
- Update Create/Edit modals with daily/weekly quota fields
- Update AccountActionMenu hasQuotaLimit to check all dimensions
- Add i18n strings for daily/weekly/total quota labels
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

抽取共享的用户分组专属倍率解析器，统一缓存、singleflight 与回退逻辑。\n\n让 OpenAI 独立计费链路复用专属倍率解析，修复 usage 记录与实际扣费未命中用户专属倍率的问题。\n\n补齐 OpenAI 计费与解析器单元测试，并修复全量回归中暴露的 lint 阻塞项。\n\nCo-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

isModelSupportedByAccount 不被 OpenAI 调度路径调用，
OpenAI /responses 和 /chat/completions 走的是
openai_account_scheduler.go，透传短路已在 PR #806 的
第二个 commit 中正确添加到该文件。

此处的检查是多余的死代码，因为 OpenAI 账号不会走到
isModelSupportedByAccount 的这个分支。

Add Anthropic Messages API support for OpenAI platform groups, enabling
clients using Claude-style /v1/messages format to access OpenAI accounts
through automatic protocol conversion.

- Add apicompat package with type definitions and bidirectional converters
  (Anthropic ↔ Chat, Chat ↔ Responses, Anthropic ↔

 Responses)
- Implement /v1/messages endpoint for OpenAI gateway with streaming support
- Add model mapping UI for OpenAI OAuth accounts (whitelist + mapping modes)
- Support prompt caching fields and codex OAuth transforms
- Fix tool call ID conversion for Responses API (fc_ prefix)
- Ensure function_call_output has non-empty output field
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

透传模式账号仅替换认证，应允许所有模型通过。之前调度阶段的
isModelSupportedByAccount 不感知透传模式，导致 model_mapping
中未配置的新模型（如 gpt-5.4）被拒绝返回 503。

- Extract postUsageBilling() to consolidate billing logic across
  GatewayService.RecordUsage, RecordUsageWithLongContext, and
  OpenAIGatewayService.RecordUsage, eliminating ~120 lines of
  duplicated code
- Fix account quota to use TotalCost × accountRateMultiplier
  (was using raw TotalCost, inconsistent with account cost stats)
- Fix RecordUsageWithLongContext API Key quota only updating in
  balance mode (now updates regardless of billing type)
- Fix WebSocket client disconnect detection on Windows by adding
  "an established connection was aborted" to known disconnect errors

- Add configurable spending limit (quota_limit) for apikey-type accounts
- Atomic quota accumulation via PostgreSQL JSONB operations on TotalCost
- Scheduler filters out over-quota accounts with outbox-triggered snapshot refresh
- Display quota usage ($used / $limit) in account capacity column
- Add "Reset Quota" action in account menu to reset usage to zero
- Editing account settings preserves quota_used (no accidental reset)
- Covers all 3 billing paths: Anthropic, Gemini, OpenAI RecordUsage

chore: bump version to 0.1.90.4