@gajae-code/ai 0.11.6 → 0.11.7

package/CHANGELOG.md

@@ -2,6 +2,12 @@
 
 ## [Unreleased]
 
+## [0.11.7] - 2026-07-22
+
+### Changed
+
+- Replaced the `alibaba-coding-plan` provider with first-class `alibaba-token-plan` support. The `/login` OAuth list, provider descriptor, model manager, models.dev descriptor, and bundled `models.json` now target the maintained Alibaba Token Plan endpoint (`https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1`, env `ALIBABA_TOKEN_PLAN_API_KEY`) and validate logins against `deepseek-v4-pro`. The retired `alibaba-coding-plan` provider pointed at `coding-intl.dashscope.aliyuncs.com`, which rejected real token-plan keys with 401 and was the only Alibaba entry exposed in `/login`.
+
 ## [0.11.4] - 2026-07-20
 
 ### Added
@@ -32,6 +38,9 @@
 ### Fixed
 
 - Fixed frequent `Request blocked (code=invalid_prompt)` failures on gpt-5.6 (Sol/Terra/Luna) subagent, default-agent, and compaction turns (ref openai/codex#32028, oh-my-pi#5184). Leaked Harmony control-token markers (e.g. `<|channel|>analysis`) were only neutralized on the replayed-history payload path, so markers in assistant reasoning summaries, live-converted message/tool-output text, and user-authored content reached the OpenAI Responses and OpenAI-codex-responses transports verbatim and wedged the session (the poisoned item was re-sent every turn). Both transports now neutralize reserved control tokens across the entire outgoing `input` array at the request boundary via an idempotent zero-width-space insertion that keeps the text human-readable.
+### Fixed
+
+- Fixed Fable 5 adaptive thinking being billed but never displayed: model discovery now classifies `claude-fable-*` as `anthropic-adaptive` (was cached as `budget`, sending `enabled`+`budget_tokens` that Fable answers with signature-only thinking), and `supportsAdaptiveThinkingDisplay` opts Fable into `display: "summarized"` on both Anthropic Messages and Bedrock Converse transports (#2791).
 
 ## [0.10.0] - 2026-07-12
 ### Fixed

package/dist/types/provider-models/openai-compat.d.ts

@@ -116,11 +116,11 @@ export interface KiloModelManagerConfig {
     baseUrl?: string;
 }
 export declare function kiloModelManagerOptions(config?: KiloModelManagerConfig): ModelManagerOptions<"openai-completions">;
-export interface AlibabaCodingPlanModelManagerConfig {
+export interface AlibabaTokenPlanModelManagerConfig {
     apiKey?: string;
     baseUrl?: string;
 }
-export declare function alibabaCodingPlanModelManagerOptions(config?: AlibabaCodingPlanModelManagerConfig): ModelManagerOptions<"openai-completions">;
+export declare function alibabaTokenPlanModelManagerOptions(config?: AlibabaTokenPlanModelManagerConfig): ModelManagerOptions<"openai-completions">;
 export interface VercelAiGatewayModelManagerConfig {
     apiKey?: string;
     baseUrl?: string;

package/dist/types/types.d.ts

@@ -51,7 +51,7 @@ export interface ThinkingConfig {
     /** Provider-specific transport used to encode the selected effort. */
     mode: ThinkingControlMode;
 }
-export type KnownProvider = "alibaba-coding-plan" | "amazon-bedrock" | "azure-openai" | "anthropic" | "google" | "google-gemini-cli" | "google-antigravity" | "google-vertex" | "openai" | "openai-codex" | "kimi-code" | "minimax-code" | "minimax-code-cn" | "github-copilot" | "fireworks" | "firepass" | "fugu" | "gitlab-duo" | "cursor" | "deepseek" | "deepinfra" | "xai" | "groq" | "cerebras" | "openrouter" | "kilo" | "vercel-ai-gateway" | "zai" | "glm-zcode" | "mistral" | "minimax" | "opencode-go" | "opencode-zen" | "synthetic" | "cloudflare-ai-gateway" | "huggingface" | "litellm" | "moonshot" | "nvidia" | "nanogpt" | "ollama" | "ollama-cloud" | "qianfan" | "qwen-portal" | "together" | "venice" | "vllm" | "xiaomi" | "xiaomi-token-plan-sgp" | "xiaomi-token-plan-ams" | "xiaomi-token-plan-cn" | "zenmux" | "lm-studio";
+export type KnownProvider = "alibaba-token-plan" | "amazon-bedrock" | "azure-openai" | "anthropic" | "google" | "google-gemini-cli" | "google-antigravity" | "google-vertex" | "openai" | "openai-codex" | "kimi-code" | "minimax-code" | "minimax-code-cn" | "github-copilot" | "fireworks" | "firepass" | "fugu" | "gitlab-duo" | "cursor" | "deepseek" | "deepinfra" | "xai" | "groq" | "cerebras" | "openrouter" | "kilo" | "vercel-ai-gateway" | "zai" | "glm-zcode" | "mistral" | "minimax" | "opencode-go" | "opencode-zen" | "synthetic" | "cloudflare-ai-gateway" | "huggingface" | "litellm" | "moonshot" | "nvidia" | "nanogpt" | "ollama" | "ollama-cloud" | "qianfan" | "qwen-portal" | "together" | "venice" | "vllm" | "xiaomi" | "xiaomi-token-plan-sgp" | "xiaomi-token-plan-ams" | "xiaomi-token-plan-cn" | "zenmux" | "lm-studio";
 export type Provider = KnownProvider | string;
 import type { Effort } from "./model-thinking";
 /** Token budgets for each thinking level (token-based providers only) */

package/dist/types/utils/oauth/alibaba-token-plan.d.ts

@@ -0,0 +1,19 @@
+/**
+ * Alibaba Token Plan login flow.
+ *
+ * Alibaba Token Plan provides OpenAI-compatible models via
+ * https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1.
+ *
+ * This is not OAuth - it's a simple API key flow:
+ * 1. Open browser to Alibaba Cloud Model Studio console
+ * 2. User copies their API key
+ * 3. User pastes the API key into the CLI
+ */
+import type { OAuthController } from "./types";
+/**
+ * Login to Alibaba Token Plan.
+ *
+ * Opens browser to API keys page, prompts user to paste their API key.
+ * Returns the API key directly (not OAuthCredentials - this isn't OAuth).
+ */
+export declare function loginAlibabaTokenPlan(options: OAuthController): Promise<string>;

package/dist/types/utils/oauth/types.d.ts

@@ -7,7 +7,7 @@ export type OAuthCredentials = {
     email?: string;
     accountId?: string;
 };
-export type OAuthProvider = "alibaba-coding-plan" | "anthropic" | "cerebras" | "cloudflare-ai-gateway" | "cursor" | "deepseek" | "deepinfra" | "fireworks" | "firepass" | "fugu" | "github-copilot" | "google-gemini-cli" | "google-antigravity" | "gitlab-duo" | "huggingface" | "kimi-code" | "kilo" | "kagi" | "litellm" | "lm-studio" | "minimax-code" | "minimax-code-cn" | "moonshot" | "nvidia" | "nanogpt" | "ollama" | "ollama-cloud" | "openai-codex" | "openai-codex-device" | "opencode-go" | "opencode-zen" | "parallel" | "perplexity" | "qianfan" | "qwen-portal" | "synthetic" | "tavily" | "together" | "venice" | "vercel-ai-gateway" | "vllm" | "xai" | "glm-zcode" | "xiaomi" | "xiaomi-token-plan-sgp" | "xiaomi-token-plan-ams" | "xiaomi-token-plan-cn" | "zenmux" | "zai";
+export type OAuthProvider = "alibaba-token-plan" | "anthropic" | "cerebras" | "cloudflare-ai-gateway" | "cursor" | "deepseek" | "deepinfra" | "fireworks" | "firepass" | "fugu" | "github-copilot" | "google-gemini-cli" | "google-antigravity" | "gitlab-duo" | "huggingface" | "kimi-code" | "kilo" | "kagi" | "litellm" | "lm-studio" | "minimax-code" | "minimax-code-cn" | "moonshot" | "nvidia" | "nanogpt" | "ollama" | "ollama-cloud" | "openai-codex" | "openai-codex-device" | "opencode-go" | "opencode-zen" | "parallel" | "perplexity" | "qianfan" | "qwen-portal" | "synthetic" | "tavily" | "together" | "venice" | "vercel-ai-gateway" | "vllm" | "xai" | "glm-zcode" | "xiaomi" | "xiaomi-token-plan-sgp" | "xiaomi-token-plan-ams" | "xiaomi-token-plan-cn" | "zenmux" | "zai";
 export type OAuthProviderId = OAuthProvider | (string & {});
 export type OAuthPrompt = {
     message: string;

package/package.json

@@ -1,7 +1,7 @@
 {
 	"type": "module",
 	"name": "@gajae-code/ai",
-	"version": "0.11.6",
+	"version": "0.11.7",
 	"description": "Unified LLM API with automatic model discovery and provider configuration",
 	"homepage": "https://gajae-code.com",
 	"author": "Yeachan-Heo and Gajae Code Contributors",
@@ -40,7 +40,7 @@
 	"dependencies": {
 		"@anthropic-ai/sdk": "^0.94.0",
 		"@bufbuild/protobuf": "^2.12.0",
-		"@gajae-code/utils": "0.11.6",
+		"@gajae-code/utils": "0.11.7",
 		"openai": "^6.36.0",
 		"partial-json": "^0.1.7",
 		"zod": "4.4.3"

package/src/auth-storage.ts

@@ -1587,9 +1587,9 @@ export class AuthStorage {
 				});
 				break;
 			}
-			case "alibaba-coding-plan": {
-				const { loginAlibabaCodingPlan } = await import("./utils/oauth/alibaba-coding-plan");
-				const apiKey = await loginAlibabaCodingPlan(ctrl);
+			case "alibaba-token-plan": {
+				const { loginAlibabaTokenPlan } = await import("./utils/oauth/alibaba-token-plan");
+				const apiKey = await loginAlibabaTokenPlan(ctrl);
 				await saveApiKeyCredential(apiKey);
 				return;
 			}

package/src/model-thinking.ts

@@ -62,7 +62,7 @@ type SemVer = {
 };
 
 type GeminiKind = "pro" | "flash";
-type AnthropicKind = "opus" | "sonnet";
+type AnthropicKind = "opus" | "sonnet" | "fable";
 type OpenAIVariant =
 	| "base"
 	| "codex"
@@ -638,6 +638,11 @@ function inferAnthropicSupportedEfforts<TApi extends Api>(
 		(model.api === "anthropic-messages" || model.api === "bedrock-converse-stream") &&
 		semverGte(parsedModel.version, "4.6")
 	) {
+		if (parsedModel.kind === "fable") {
+			// Fable exposes Anthropic's Messages-only xhigh preset; Bedrock
+			// Converse lacks it (same split as Opus 4.7+ below).
+			return model.api === "anthropic-messages" ? DEFAULT_REASONING_EFFORTS_WITH_XHIGH : DEFAULT_REASONING_EFFORTS;
+		}
 		if (parsedModel.kind !== "opus") return DEFAULT_REASONING_EFFORTS;
 		return anthropicModelHasRealXHighEffort(model)
 			? DEFAULT_REASONING_EFFORTS_WITH_XHIGH_AND_MAX
@@ -697,7 +702,10 @@ function inferThinkingControlMode<TApi extends Api>(
 
 		case "bedrock-converse-stream":
 			if (parsedModel.family === "anthropic") {
-				if (semverGte(parsedModel.version, "4.6") && parsedModel.kind === "opus") {
+				if (
+					semverGte(parsedModel.version, "4.6") &&
+					(parsedModel.kind === "opus" || parsedModel.kind === "fable")
+				) {
 					return "anthropic-adaptive";
 				}
 				if (semverGte(parsedModel.version, "4.5")) {
@@ -737,7 +745,7 @@ function parseGeminiModel(modelId: string): GeminiModel | null {
 }
 
 function parseAnthropicModel(modelId: string): AnthropicModel | null {
-	const match = /claude-(opus|sonnet)-(\d{1,2}(?:[.-]\d{1,2}){0,2})\b/.exec(modelId);
+	const match = /claude-(opus|sonnet|fable)-(\d{1,2}(?:[.-]\d{1,2}){0,2})\b/.exec(modelId);
 	if (!match) {
 		return null;
 	}