github lbjlaq/Antigravity-Manager v5.0.4-beta.15
Antigravity Tools v5.0.4-beta.15

pre-releaseone hour ago
  • [Quota Endpoint Host Alignment & Exhausted Model 0% Fallback Fix] (Fixes #3987, PR #3988, PR #3989, Thanks to @vecyang1):
    • Drop Prod Host from Quota Endpoints to Match Inference: Removed cloudcode-pa.googleapis.com (Prod) from QUOTA_API_ENDPOINTS and QUOTA_SUMMARY_ENDPOINTS so quota polling exclusively uses Daily -> Sandbox alongside inference (V1_INTERNAL_BASE_URL_FALLBACKS), guarded by a host-subset invariant unit test; eliminates 100% / 0% Gemini quota flip-flopping and scheduling of depleted accounts into 429 QUOTA_EXHAUSTED upon transient Daily failures. (Fixes #3987, PR #3988, Thanks to @vecyang1)
    • Treat Omitted remainingFraction with Active resetTime as 0%: Extracted model_quota_percentage so quotaInfo entries carrying a non-empty resetTime without remainingFraction (omitted by Protobuf v3 zero-value serialization) resolve to Some(0) while keeping missing/empty quotaInfo on group-quota fallback, preventing exhausted models from displaying 100% and losing resetTime if retrieveUserQuotaSummary fails concurrently. (Ref #3987, PR #3989, Thanks to @vecyang1)
  • [Same-Account Auto-Heal Clamping for Unsupported Extended Thinking Levels] (Fixes #3991, PR #3992, Thanks to @vecyang1):
    • Clamp Extended Thinking Level to HIGH on 400 Thinking level is unsupported: Added a single-retry same-account auto-heal arm (try_clamp_thinking_level) in the Pipeline for accounts that advertise -xhigh (thinkingLevel: 100) in fetchAvailableModels but reject it at generation time with 400 INVALID_ARGUMENT: Thinking level is unsupported: THINKING_LEVEL_XHIGH. Clamps extended levels (> 3 or XHIGH) to HIGH and downgrades -xhigh model IDs to -high across all four protocol handlers without breaking sticky-session KV Cache or tripping family-wide Hydrogen circuit breakers. (Fixes #3991, PR #3992, Thanks to @vecyang1)
  • [OpenAI Adapter Sanitization Timing Fix & Codex Persona Normalization] (Ref #3981, Fixes #3982, Ref #3983, Ref #3984, PR #3986, Thanks to @cubelikeplayDaniel):
    • Relocate sanitize_gemini_payload in OpenAI Request Mapper: Moved payload sanitization in the OpenAI request mapper to run after contents, tools, systemInstruction, and inject_google_search_tool are fully populated and before prefix topology alignment and stealth sideloading (aligned with the Claude adapter), preventing premature sideloading of unnormalized system prompts, duplicate identity blocks, and stale L3-Prefix hashes while preserving L2-Tools cache hot-reload support. (Ref #3981, Ref #3983, Ref #3984, PR #3986, Thanks to @cubelikeplayDaniel)
    • Normalize Short Codex Client Declarations: Extended RE_KNOWN_CLIENT_DECLARATION to match standalone codex assistant/agent preambles without vendor or model attribution clauses (e.g. You are Codex, an AI coding assistant.), normalizing them into official Antigravity native identity. (Fixes #3982, PR #3986, Thanks to @cubelikeplayDaniel)
    • Harden Concurrent Unit Test Config Isolation: Added TEST_CONFIG_LOCK synchronization and explicit default configuration initialization/reset across prompt sanitizer and OpenAI mapper tests, alongside direct regression assertions for tool description sanitization and underscore identifier preservation. (Ref #3981, PR #3986, Thanks to @cubelikeplayDaniel)

  • [配额查询端点对齐与耗尽模型 0% 兜底判定修复] (Fixes #3987, PR #3988, PR #3989, Thanks to @vecyang1):
    • 配额接口移除 Prod 端点对齐推理链路: 从 QUOTA_API_ENDPOINTS 与 QUOTA_SUMMARY_ENDPOINTS 中移除推理链路早已弃用的 cloudcode-pa.googleapis.com(Prod),统一收敛为 Daily -> Sandbox 并新增跨模块主机集合一致性单测断言,根治 Daily 瞬态波动回退至 Prod 读取失真 Gemini 配额导致面板额度在 100% / 0% 间反复跳变及误调度已耗尽账号触发 429 QUOTA_EXHAUSTED 的问题。(Fixes #3987, PR #3988, Thanks to @vecyang1)
    • 识别 Protobuf 省略零值字段的已耗尽 quotaInfo: 新增 model_quota_percentage 解析逻辑,当 fetchAvailableModels 下发的 quotaInfo 携带非空 resetTime 但因 Protobuf v3 默认零值省略机制缺失 remainingFraction 时精准判定为 0%(对缺失 quotaInfo 或无有效 resetTime 的模型仍维持回退分组配额),防止同轮 retrieveUserQuotaSummary 失败时将已耗尽模型误判为 100% 并丢失重置时间。(Ref #3987, PR #3989, Thanks to @vecyang1)
  • [扩展思考档位不支持(400 XHIGH)同账号降档自愈] (Fixes #3991, PR #3992, Thanks to @vecyang1):
    • 400 Thinking level is unsupported 单次钳制 HIGH 原地重试: 针对部分账号虽在目录中声明支持 gemini-4-hydrogen-xhigh(thinkingLevel: 100)但推理时返回 400 INVALID_ARGUMENT: Thinking level is unsupported: THINKING_LEVEL_XHIGH 的问题,在 Pipeline try_auto_heal 中新增同账号单次自愈分支(try_clamp_thinking_level),将扩展档位(数值 > 3 或 XHIGH)安全钳制为 HIGH 并将 -xhigh 模型名同步降为 -high 后原地重试(四协议入口重建请求时同步保持钳制状态),既避免整轮对话失败与跨账号丢失 KV Cache,也避免触发模型熔断连坐冻结全系 Hydrogen 档位。(Fixes #3991, PR #3992, Thanks to @vecyang1)
  • [OpenAI 适配器请求清洗时序修正与 Codex 身份拟态补齐] (Ref #3981, Fixes #3982, Ref #3983, Ref #3984, PR #3986, Thanks to @cubelikeplayDaniel):
    • 修正 OpenAI 适配器 sanitize_gemini_payload 执行时序: 将 OpenAI 兼容请求映射器中的提示词与工具描述清洗节点下移至 contents、tools、systemInstruction 及 inject_google_search_tool 全量装配完成之后、前缀拓扑对齐与隐身侧载执行之前(与 Claude 适配器一致),消除隐身侧载引擎与 L3-Prefix 缓存哈希读取未归一化原文导致的误侧载及重复身份隐患,同时保留 L2-Tools 缓存的规则热更新能力。(Ref #3981, Ref #3983, Ref #3984, PR #3986, Thanks to @cubelikeplayDaniel)
    • 补齐 Codex 短开场白原生身份拟态: 在 RE_KNOWN_CLIENT_DECLARATION 中增补独立 codex 助手声明匹配分支,将不携带厂牌或模型归属后缀的短开场白(如 You are Codex, an AI coding assistant.)无缝归一化为官方 Antigravity 原生身份。(Fixes #3982, PR #3986, Thanks to @cubelikeplayDaniel)
    • 并发单测配置锁隔离与边界断言加固: 为提示词清洗引擎与 OpenAI 映射器单元测试补齐 TEST_CONFIG_LOCK 互斥锁及默认配置初始化复位,补充工具描述脱敏与下划线标识符保留的直接回归测试。(Ref #3981, PR #3986, Thanks to @cubelikeplayDaniel)

What's Changed

  • fix(proxy): restore MCP namespace fields for Codex Responses (Fixes #3979) / 修复 Codex Responses 的 MCP namespace 字段 by @Lucas-Qh-Lai in #3980
  • fix(oauth): request aicode scope for AGY CLI 1.5.x by @ThermalEng in #3978
  • fix(proxy): add full support for Antigravity GCP PAYGO tier and restore project ID commands (Fixes #3960) by @cubelikeplayDaniel in #3970
  • fix(proxy): fix OpenAI adapter sanitization timing and normalize Codex client persona (Fixes #3981, #3982, #3983, Ref #3984) by @cubelikeplayDaniel in #3986
  • fix(quota): 配额端点移除 Prod,与推理端点保持一致 (Fixes #3987) by @vecyang1 in #3988
  • fix(quota): quotaInfo 省略 remainingFraction 时按 0 处理,避免耗尽模型显示 100% (Ref #3987) by @vecyang1 in #3989
  • fix(proxy): 400 "Thinking level is unsupported" 时同账号钳制为 HIGH 重试 (Fixes #3991) by @vecyang1 in #3992

New Contributors

Full Changelog: v5.0.4-beta.14...v5.0.4-beta.15

Don't miss a new Antigravity-Manager release

NewReleases is sending notifications on new releases.