@fudrouter/fsrouter 0.6.72 → 0.6.74

package/open-sse/config/appConstants.js

@@ -1,44 +1,10 @@
 import { platform, arch } from "os";
-import { proxyAwareFetch } from "../utils/proxyFetch.js";
-
-// === Kimchi CLI version (dynamic, cached from GitHub releases) ===
-const KIMCHI_FALLBACK_VERSION = "0.1.39";
-let _kimchiVersionCache = null;
-let _kimchiVersionFetching = null;
-const KIMCHI_VERSION_TTL_MS = 3600_000; // 1h
-
-async function fetchKimchiVersion() {
-  try {
-    const res = await proxyAwareFetch("https://api.github.com/repos/getkimchi/kimchi/releases/latest", {
-      headers: { Accept: "application/vnd.github+json" },
-      signal: AbortSignal.timeout(5000),
-    });
-    if (!res.ok) return KIMCHI_FALLBACK_VERSION;
-    const body = await res.json();
-    const tag = body?.tag_name || "";
-    const ver = tag.replace(/^v/, "");
-    return ver || KIMCHI_FALLBACK_VERSION;
-  } catch {
-    return KIMCHI_FALLBACK_VERSION;
-  }
-}
-
-export async function getKimchiVersion() {
-  if (_kimchiVersionCache && Date.now() - _kimchiVersionCache.ts < KIMCHI_VERSION_TTL_MS) return _kimchiVersionCache.ver;
-  if (_kimchiVersionFetching) return _kimchiVersionFetching;
-  _kimchiVersionFetching = fetchKimchiVersion().finally(() => { _kimchiVersionFetching = null; });
-  const ver = await _kimchiVersionFetching;
-  _kimchiVersionCache = { ver, ts: Date.now() };
-  return ver;
-}
+import { PROVIDERS, PROVIDER_OAUTH } from "./providers.js";
+import { ANTIGRAVITY_IDE_USER_AGENT } from "../providers/shared.js";
 
-export function getKimchiVersionSync() {
-  return _kimchiVersionCache?.ver || KIMCHI_FALLBACK_VERSION;
-}
-
-// === Gemini CLI ===
-export const GEMINI_CLI_VERSION = "0.34.0";
-export const GEMINI_CLI_API_CLIENT = "google-genai-sdk/1.41.0 gl-node/v22.19.0";
+// === Gemini CLI === derive từ registry gemini-cli.transport
+export const GEMINI_CLI_VERSION = PROVIDERS["gemini-cli"]?.cliVersion;
+export const GEMINI_CLI_API_CLIENT = PROVIDERS["gemini-cli"]?.apiClient;
 
 // Map Node arch to Gemini CLI arch string (x64/x86/arm64/...)
 function geminiCLIArch() {
@@ -52,11 +18,13 @@ export function geminiCLIUserAgent(model = "unknown") {
 }
 
 // === GitHub Copilot ===
+// Derive từ registry github.transport.copilot
+const _ghCopilot = PROVIDERS.github?.copilot || {};
 export const GITHUB_COPILOT = {
-  VSCODE_VERSION: "1.110.0",
-  COPILOT_CHAT_VERSION: "0.38.0",
-  USER_AGENT: "GitHubCopilotChat/0.38.0",
-  API_VERSION: "2025-04-01",
+  VSCODE_VERSION: _ghCopilot.vscodeVersion,
+  COPILOT_CHAT_VERSION: _ghCopilot.chatVersion,
+  USER_AGENT: _ghCopilot.userAgent,
+  API_VERSION: _ghCopilot.apiVersion,
 };
 
 // === Antigravity enums ===
@@ -92,7 +60,7 @@ export function getPlatformEnum() {
 }
 
 export function getPlatformUserAgent() {
-  return `antigravity/1.104.0 ${platform()}/${arch()}`;
+  return ANTIGRAVITY_IDE_USER_AGENT;
 }
 
 export const CLIENT_METADATA = {
@@ -162,7 +130,7 @@ export const AG_DEFAULT_TOOLS = new Set([
 
 // Antigravity chat/stream headers
 export const ANTIGRAVITY_HEADERS = {
-  "User-Agent": `antigravity/1.107.0 ${platform()}/${arch()}`
+  "User-Agent": ANTIGRAVITY_IDE_USER_AGENT
 };
 
 // Cloud Code Assist API
@@ -188,43 +156,19 @@ export const LOAD_CODE_ASSIST_METADATA = {
 export const CLAUDE_SYSTEM_PROMPT = "You are Claude Code, Anthropic's official CLI for Claude.";
 export const ANTIGRAVITY_DEFAULT_SYSTEM = "You are Antigravity, a powerful agentic AI coding assistant designed by the Google Deepmind team working on Advanced Agentic Coding.You are pair programming with a USER to solve their coding task. The task may require creating a new codebase, modifying or debugging an existing codebase, or simply answering a question.**Absolute paths only****Proactiveness**";
 
-// Proactive token refresh lead times per provider (ms)
-export const REFRESH_LEAD_MS = {
-  codex:       5 * 24 * 60 * 60 * 1000, // 5 days
-  claude:       4 * 60 * 60 * 1000,     // 4 hours
-  iflow:       24 * 60 * 60 * 1000,     // 24 hours
-  qwen:        20 * 60 * 1000,          // 20 minutes
-  "kimi-coding": 5 * 60 * 1000,         // 5 minutes
-  antigravity:  5 * 60 * 1000,          // 5 minutes
-};
+// Derive từ registry oauth.refreshLeadMs
+export const REFRESH_LEAD_MS = Object.fromEntries(
+  Object.entries(PROVIDER_OAUTH).filter(([, o]) => o.refreshLeadMs).map(([id, o]) => [id, o.refreshLeadMs])
+);
 
 // OAuth endpoints
 export const OAUTH_ENDPOINTS = {
-  google: {
-    token: "https://oauth2.googleapis.com/token",
-    auth: "https://accounts.google.com/o/oauth2/auth"
-  },
-  openai: {
-    token: "https://auth.openai.com/oauth/token",
-    auth: "https://auth.openai.com/oauth/authorize"
-  },
-  anthropic: {
-    token: "https://api.anthropic.com/v1/oauth/token",
-    auth: "https://api.anthropic.com/v1/oauth/authorize"
-  },
-  qwen: {
-    token: "https://qwen.ai/api/v1/oauth2/token",
-    auth: "https://qwen.ai/api/v1/oauth2/device/code"
-  },
-  iflow: {
-    token: "https://iflow.cn/oauth/token",
-    auth: "https://iflow.cn/oauth"
-  },
-  github: {
-    token: "https://github.com/login/oauth/access_token",
-    auth: "https://github.com/login/oauth/authorize",
-    deviceCode: "https://github.com/login/device/code"
-  }
+  google:    { token: "https://oauth2.googleapis.com/token", auth: "https://accounts.google.com/o/oauth2/auth" },
+  openai:    { token: PROVIDER_OAUTH["codex"]?.tokenUrl, auth: PROVIDER_OAUTH["codex"]?.authorizeUrl },
+  anthropic: { token: PROVIDER_OAUTH["claude"]?.tokenUrl, auth: "https://api.anthropic.com/v1/oauth/authorize" }, // ≠ claude.authorizeUrl (claude.ai login) — keep
+  qwen:      { token: PROVIDER_OAUTH["qwen"]?.tokenUrl, auth: PROVIDER_OAUTH["qwen"]?.deviceCodeUrl },
+  iflow:     { token: PROVIDER_OAUTH["iflow"]?.tokenUrl, auth: PROVIDER_OAUTH["iflow"]?.authorizeUrl },
+  github:    { token: PROVIDER_OAUTH["github"]?.tokenUrl, auth: PROVIDER_OAUTH["github"]?.authorizeUrl, deviceCode: PROVIDER_OAUTH["github"]?.deviceCodeUrl },
 };
 
 // Generate Kimi OAuth custom headers

package/open-sse/config/errorConfig.js

@@ -45,7 +45,6 @@ export const MAX_RATE_LIMIT_COOLDOWN_MS = 30 * 60 * 1000;
 const COOLDOWN = {
   long: 2 * 60 * 1000,
   short: 5 * 1000,
-  exhausted: 24 * 60 * 60 * 1000, // 24 hours
 };
 
 /**
@@ -59,16 +58,6 @@ const COOLDOWN = {
  */
 export const ERROR_RULES = [
   // --- Text-based rules (checked first, order = priority) ---
-  { text: "exhausted",                cooldownMs: COOLDOWN.exhausted },
-  { text: "insufficient credits",     cooldownMs: COOLDOWN.exhausted },
-  { text: "1076",                     cooldownMs: COOLDOWN.exhausted },
-  { text: "only available on paid plans", cooldownMs: COOLDOWN.exhausted },
-  { text: "failed to capture weavy firebase id token", cooldownMs: COOLDOWN.exhausted },
-  { text: "trial version is not yet", cooldownMs: COOLDOWN.exhausted },
-  { text: "neurons",                  cooldownMs: COOLDOWN.exhausted }, // CF Workers AI daily quota
-  { text: "daily free allocation",    cooldownMs: COOLDOWN.exhausted }, // CF Workers AI daily quota
-  { text: "upgrade to cloudflare",    cooldownMs: COOLDOWN.exhausted }, // CF Workers AI paid plan prompt
-  { text: "4006",                     cooldownMs: COOLDOWN.exhausted }, // CF Workers AI error code
   { text: "no credentials",           cooldownMs: COOLDOWN.long },
   { text: "request not allowed",      cooldownMs: COOLDOWN.short },
   { text: "improperly formed request", cooldownMs: COOLDOWN.long },

package/open-sse/config/grokCli.js

@@ -0,0 +1,10 @@
+export const GROK_CLI_VERSION = "0.2.99";
+export const GROK_CLI_MODEL = "grok-build";
+export const GROK_CLI_BASE_URL = "https://cli-chat-proxy.grok.com/v1";
+export const GROK_CLI_CLIENT_IDENTIFIER = "grok-shell";
+export const GROK_CLI_USER_AGENT = `grok-shell/${GROK_CLI_VERSION} (linux; x86_64)`;
+
+export function supportsGrokCliReasoningEffort(model) {
+  // ponytail: unknown models omit effort until live metadata reaches dispatch.
+  return /^grok-4\.5(?:$|-)/.test(String(model || ""));
+}

package/open-sse/config/kiroConstants.js

@@ -15,9 +15,29 @@
  * fiction. The suffix is stripped before the request leaves this process.
  */
 
+import { extractThinking } from "../translator/concerns/thinkingUnified.js";
+import { effortToBudget } from "../translator/concerns/thinking.js";
+
 export const KIRO_AGENTIC_SUFFIX = "-agentic";
 export const KIRO_THINKING_SUFFIX = "-thinking";
 
+// Public default CodeWhisperer profile ARNs (us-east-1), keyed by auth method.
+// Used when an account cannot resolve its own profileArn. Builder ID and social
+// (Google/GitHub) sign-ins map to different shared profiles.
+export const KIRO_DEFAULT_PROFILE_ARNS = {
+  "builder-id": "arn:aws:codewhisperer:us-east-1:638616132270:profile/AAAACCCCXXXX",
+  social: "arn:aws:codewhisperer:us-east-1:699475941385:profile/EHGA3GRVQMUK",
+};
+
+// Back-compat single default (Builder ID).
+export const KIRO_DEFAULT_PROFILE_ARN = KIRO_DEFAULT_PROFILE_ARNS["builder-id"];
+
+/** Resolve the shared default profileArn for a given auth method. */
+export function resolveDefaultProfileArn(authMethod) {
+  const social = authMethod === "google" || authMethod === "github";
+  return social ? KIRO_DEFAULT_PROFILE_ARNS.social : KIRO_DEFAULT_PROFILE_ARNS["builder-id"];
+}
+
 export const KIRO_THINKING_BUDGET_DEFAULT = 16000;
 
 export const KIRO_AGENTIC_SYSTEM_PROMPT = `
@@ -72,61 +92,100 @@ REMEMBER: When in doubt, write LESS per operation. Multiple small operations > o
 `.trim();
 
 /**
- * Detect whether an inbound request is asking for reasoning / thinking output.
+ * Resolve the Kiro thinking budget requested by a client.
  *
- * Sources of intent (any one is enough):
- *   - HTTP header `Anthropic-Beta: ...interleaved-thinking...`
- *   - JSON `thinking.type === "enabled"` (Claude Messages API)
- *   - JSON `reasoning_effort` in {low, medium, high, auto} (OpenAI o1/o3)
- *   - JSON `reasoning.effort` in {low, medium, high, auto} (OpenAI Responses)
- *   - System prompt contains `<thinking_mode>enabled</thinking_mode>` or
- *     `<thinking_mode>interleaved</thinking_mode>` (AMP / Cursor)
- *   - Model name contains `thinking` or `-reason`
+ * Reuses the shared thinkingUnified parser (extractThinking) so every client
+ * shape (Claude output_config.effort / thinking.budget_tokens, OpenAI
+ * reasoning_effort / reasoning.effort, Gemini, Qwen) maps consistently. Explicit
+ * `none`/`off`/disabled wins and returns null (no prefix injected).
+ * buildThinkingSystemPrefix performs Kiro's final 1..32000 clamp.
  *
- * @param {object} body OpenAI-shaped request body (post-translation)
+ * @param {object} body OpenAI/Claude-shaped request body
  * @param {object} [headers] Original inbound HTTP headers (case-insensitive)
- * @param {string} [model] Model id the caller asked for (post-strip ok)
- * @returns {boolean}
+ * @param {string} [model] Model id the caller asked for
+ * @returns {number|null} budget to inject, or null when thinking is disabled
  */
-export function isThinkingEnabled(body, headers, model) {
+export function resolveKiroThinkingBudget(body, headers, model) {
+  const cfg = extractThinking(body);
+  if (cfg) {
+    if (cfg.mode === "none") return null;
+    if (cfg.mode === "budget") return cfg.budget;
+    if (cfg.mode === "level") return effortToBudget(cfg.level) ?? KIRO_THINKING_BUDGET_DEFAULT;
+    return KIRO_THINKING_BUDGET_DEFAULT;
+  }
+
   if (headers) {
     const beta = pickHeader(headers, "anthropic-beta");
     if (typeof beta === "string" && beta.toLowerCase().includes("interleaved-thinking")) {
-      return true;
+      return KIRO_THINKING_BUDGET_DEFAULT;
     }
   }
 
-  if (body && typeof body === "object") {
-    const thinking = body.thinking;
-    if (thinking && typeof thinking === "object" && thinking.type === "enabled") {
-      const budget = Number(thinking.budget_tokens);
-      if (!Number.isFinite(budget) || budget > 0) {
-        return true;
-      }
-    }
-
-    const effort = body.reasoning_effort
-      ?? (body.reasoning && typeof body.reasoning === "object" ? body.reasoning.effort : null);
-    if (typeof effort === "string") {
-      const v = effort.toLowerCase();
-      if (v && v !== "none" && (v === "low" || v === "medium" || v === "high" || v === "auto")) {
-        return true;
-      }
-    }
-
-    if (containsThinkingModeTag(body)) {
-      return true;
-    }
-  }
+  if (containsThinkingModeTag(body)) return KIRO_THINKING_BUDGET_DEFAULT;
 
   if (typeof model === "string" && model) {
     const m = model.toLowerCase();
-    if (m.includes("thinking") || m.includes("-reason")) {
-      return true;
-    }
+    if (m.includes("thinking") || m.includes("-reason")) return KIRO_THINKING_BUDGET_DEFAULT;
   }
 
-  return false;
+  return null;
+}
+
+export function extractKiroEffortLevel(body) {
+  const effort =
+    body?.output_config?.effort ??
+    body?.reasoning_effort ??
+    (typeof body?.reasoning === "object" ? body.reasoning?.effort : null);
+  if (typeof effort !== "string") return null;
+  const normalized = effort.toLowerCase();
+  if (normalized === "none" || normalized === "off" || normalized === "disabled") return null;
+  if (normalized === "xhigh" || normalized === "max") return "high";
+  if (["low", "medium", "high"].includes(normalized)) return normalized;
+  return null;
+}
+
+export function buildKiroAdditionalModelRequestFields(body) {
+  const effort = extractKiroEffortLevel(body);
+  if (!effort) return undefined;
+  // Mirrors Kiro CLI/KAS buildEffortRequestFields("output_config").
+  return {
+    thinking: { type: "adaptive", display: "summarized" },
+    output_config: { effort },
+  };
+}
+
+export function supportsKiroAdditionalModelRequestFields(model) {
+  if (typeof model !== "string") return false;
+  const normalized = model.toLowerCase().replace(/-/g, ".");
+  if (!normalized.includes("claude")) return false;
+  const match = normalized.match(/(?:^|[/.])claude(?:[/.][a-z]+)*[/.](\d+)(?:[/.](\d+))?(?:[/.]|$)/);
+  if (!match) return false;
+  const [, majorText, minorText] = match;
+  const major = Number(majorText);
+  const minor = minorText === undefined ? null : Number(minorText);
+  const dateSuffixMinor = minor !== null && minor >= 1000;
+  // Kiro rejected additionalModelRequestFields on legacy 4.5 models in live smoke.
+  // Default future Claude/Kiro models to supported so new model releases do not
+  // need a code allowlist update.
+  return !(major < 4 || (major === 4 && (minor === null || minor <= 5 || dateSuffixMinor)));
+}
+
+export function buildKiroAdditionalModelRequestFieldsForModel(body, model) {
+  if (!supportsKiroAdditionalModelRequestFields(model)) return undefined;
+  return buildKiroAdditionalModelRequestFields(body);
+}
+
+/**
+ * Detect whether an inbound request is asking for reasoning / thinking output.
+ * Thin wrapper over resolveKiroThinkingBudget (single source of truth).
+ *
+ * @param {object} body OpenAI-shaped request body (post-translation)
+ * @param {object} [headers] Original inbound HTTP headers (case-insensitive)
+ * @param {string} [model] Model id the caller asked for (post-strip ok)
+ * @returns {boolean}
+ */
+export function isThinkingEnabled(body, headers, model) {
+  return resolveKiroThinkingBudget(body, headers, model) !== null;
 }
 
 /**
@@ -260,9 +319,3 @@ function containsTagInText(text) {
   return text.includes("<thinking_mode>enabled</thinking_mode>")
     || text.includes("<thinking_mode>interleaved</thinking_mode>");
 }
-
-// Added from 9router - missing exports
-export function resolveDefaultProfileArn(authMethod) {
-  const social = authMethod === "google" || authMethod === "github";
-  return social ? KIRO_DEFAULT_PROFILE_ARNS.social : KIRO_DEFAULT_PROFILE_ARNS["builder-id"];
-}

package/open-sse/config/runtimeConfig.js

@@ -65,6 +65,8 @@ export const GEMINI_NATIVE_TTS_FETCH_TIMEOUT_MS = envMs("GEMINI_NATIVE_TTS_FETCH
 export const DEFAULT_MAX_TOKENS = 64000;
 export const DEFAULT_MIN_TOKENS = 32000;
 
+export const TOKEN_SAVER_HEADER = "x-9router-token-saver";
+
 // Retry config for 429 responses (legacy - kept for backward compatibility)
 export const RETRY_CONFIG = {
   maxAttempts: 2,

package/open-sse/config/ttsModels.js

@@ -96,10 +96,12 @@ export const TTS_MODELS_CONFIG = {
   },
   gemini: {
     models: [
+      { id: "gemini-3.1-flash-tts-preview", name: "Gemini 3.1 Flash TTS", type: "tts" },
       { id: "gemini-2.5-flash-preview-tts", name: "Gemini 2.5 Flash TTS", type: "tts" },
       { id: "gemini-2.5-pro-preview-tts",   name: "Gemini 2.5 Pro TTS",   type: "tts" },
     ],
     voices: {
+      "gemini-3.1-flash-tts-preview": GEMINI_VOICES,
       "gemini-2.5-flash-preview-tts": GEMINI_VOICES,
       "gemini-2.5-pro-preview-tts":   GEMINI_VOICES,
     },