@objectstack/service-settings 6.9.0 → 7.1.0

package/dist/index.cjs

@@ -907,8 +907,8 @@ var manifest = {
   icon: "Mail",
   description: "SMTP and transactional email provider configuration.",
   scope: "global",
-  readPermission: "setup.access",
-  writePermission: "setup.write",
+  readPermission: "manage_platform_settings",
+  writePermission: "manage_platform_settings",
   category: "Communication",
   order: 10,
   specifiers: [
@@ -1121,8 +1121,8 @@ var manifest2 = {
   icon: "HardDrive",
   description: "Backend used for attachments, exports, and user uploads. \u26A0 Switching adapter does not migrate existing files \u2014 files uploaded under the previous adapter become unreachable through the new one.",
   scope: "global",
-  readPermission: "setup.access",
-  writePermission: "setup.write",
+  readPermission: "manage_platform_settings",
+  writePermission: "manage_platform_settings",
   category: "Infrastructure",
   order: 20,
   specifiers: [
@@ -1294,8 +1294,8 @@ var manifest3 = {
   icon: "Sparkles",
   description: "LLM provider, model, credentials, and embedder configuration used by the platform AI and knowledge services. Provider SDK packages (e.g. @ai-sdk/openai for chat, @objectstack/embedder-openai for embeddings) must be installed on the host for the chosen provider to be loadable at runtime.",
   scope: "global",
-  readPermission: "setup.access",
-  writePermission: "setup.write",
+  readPermission: "manage_platform_settings",
+  writePermission: "manage_platform_settings",
   category: "Infrastructure",
   order: 30,
   specifiers: [
@@ -1318,7 +1318,16 @@ var manifest3 = {
         { value: "gateway", label: "Vercel AI Gateway" },
         { value: "openai", label: "OpenAI" },
         { value: "anthropic", label: "Anthropic" },
-        { value: "google", label: "Google Generative AI" }
+        { value: "google", label: "Google Generative AI" },
+        // Below: providers expose an OpenAI-compatible Chat Completions API,
+        // so they reuse the @ai-sdk/openai SDK with a preset base_url. The
+        // plugin's buildAdapterFromValues maps these onto provider="openai"
+        // + an auto-filled openai_base_url at runtime.
+        { value: "deepseek", label: "DeepSeek (OpenAI-compatible)" },
+        { value: "dashscope", label: "\u963F\u91CC\u901A\u4E49 DashScope (OpenAI-compatible)" },
+        { value: "cloudflare", label: "Cloudflare AI Gateway (OpenAI-compatible)" },
+        { value: "siliconflow", label: "\u7845\u57FA\u6D41\u52A8 SiliconFlow (OpenAI-compatible)" },
+        { value: "openrouter", label: "OpenRouter (OpenAI-compatible)" }
       ]
     },
     // ── Vercel AI Gateway ─────────────────────────────────────────
@@ -1431,6 +1440,160 @@ var manifest3 = {
       default: "gemini-2.0-flash",
       visible: "${data.provider === 'google'}"
     },
+    // ── OpenAI-compatible presets (DeepSeek / DashScope / Cloudflare / …) ──
+    //
+    // These providers all expose `/v1/chat/completions` in OpenAI shape, so
+    // we reuse the `@ai-sdk/openai` SDK with a preset base URL. The plugin
+    // normalises `provider=deepseek` (etc.) to `provider=openai` at adapter
+    // construction time, injecting the right base URL and a sensible default
+    // model id. Users only fill in API key + (optionally) model — the URL
+    // is preset, eliminating the #1 onboarding mistake.
+    {
+      type: "group",
+      id: "deepseek",
+      label: "DeepSeek",
+      required: false,
+      visible: "${data.provider === 'deepseek'}",
+      description: "OpenAI-compatible API at https://api.deepseek.com. Base URL is auto-filled."
+    },
+    {
+      type: "password",
+      key: "deepseek_api_key",
+      label: "DeepSeek API key",
+      required: true,
+      encrypted: true,
+      description: "sk-... \u2014 issued at platform.deepseek.com.",
+      visible: "${data.provider === 'deepseek'}"
+    },
+    {
+      type: "text",
+      key: "deepseek_model",
+      label: "Model",
+      required: false,
+      default: "deepseek-chat",
+      description: "Examples: deepseek-chat (V3), deepseek-reasoner (R1 thinking).",
+      visible: "${data.provider === 'deepseek'}"
+    },
+    {
+      type: "group",
+      id: "dashscope",
+      label: "\u963F\u91CC\u901A\u4E49 DashScope",
+      required: false,
+      visible: "${data.provider === 'dashscope'}",
+      description: "OpenAI-compatible endpoint at dashscope.aliyuncs.com/compatible-mode/v1. Base URL is auto-filled."
+    },
+    {
+      type: "password",
+      key: "dashscope_api_key",
+      label: "DashScope API key",
+      required: true,
+      encrypted: true,
+      description: "sk-... \u2014 issued at dashscope.console.aliyun.com.",
+      visible: "${data.provider === 'dashscope'}"
+    },
+    {
+      type: "text",
+      key: "dashscope_model",
+      label: "Model",
+      required: false,
+      default: "qwen-plus",
+      description: "Examples: qwen-plus, qwen-max, qwen3-max, qwen-turbo.",
+      visible: "${data.provider === 'dashscope'}"
+    },
+    {
+      type: "group",
+      id: "cloudflare",
+      label: "Cloudflare AI Gateway",
+      required: false,
+      visible: "${data.provider === 'cloudflare'}",
+      description: "Uses the /compat endpoint so the model id is `provider/model` (e.g. `openai/gpt-4o-mini`, `anthropic/claude-3-5-sonnet`, `deepseek/deepseek-chat`). Note: alibaba/qwen* is NOT supported by Cloudflare /compat \u2014 use the DashScope provider for Qwen."
+    },
+    {
+      type: "text",
+      key: "cloudflare_account_id",
+      label: "Cloudflare account id",
+      required: true,
+      description: "The 32-char hex id from your Cloudflare dashboard URL.",
+      visible: "${data.provider === 'cloudflare'}"
+    },
+    {
+      type: "text",
+      key: "cloudflare_gateway_id",
+      label: "Gateway id",
+      required: false,
+      default: "default",
+      description: "Gateway name configured in Cloudflare \u2192 AI Gateway. Defaults to `default`.",
+      visible: "${data.provider === 'cloudflare'}"
+    },
+    {
+      type: "password",
+      key: "cloudflare_api_key",
+      label: "Cloudflare AI Gateway token",
+      required: true,
+      encrypted: true,
+      description: 'Issued in AI Gateway \u2192 "API tokens" tab (cfut_\u2026 or sk_\u2026).',
+      visible: "${data.provider === 'cloudflare'}"
+    },
+    {
+      type: "text",
+      key: "cloudflare_model",
+      label: "Model",
+      required: false,
+      default: "openai/gpt-4o-mini",
+      description: "Format: provider/model. Allowed providers (per Cloudflare /compat): anthropic, openai, groq, mistral, cohere, perplexity, workers-ai, google-ai-studio, vertex, grok, deepseek, cerebras, baseten, parallel.",
+      visible: "${data.provider === 'cloudflare'}"
+    },
+    {
+      type: "group",
+      id: "siliconflow",
+      label: "\u7845\u57FA\u6D41\u52A8 SiliconFlow",
+      required: false,
+      visible: "${data.provider === 'siliconflow'}",
+      description: "OpenAI-compatible endpoint at api.siliconflow.cn/v1. Base URL is auto-filled."
+    },
+    {
+      type: "password",
+      key: "siliconflow_api_key",
+      label: "SiliconFlow API key",
+      required: true,
+      encrypted: true,
+      visible: "${data.provider === 'siliconflow'}"
+    },
+    {
+      type: "text",
+      key: "siliconflow_model",
+      label: "Model",
+      required: false,
+      default: "Qwen/Qwen2.5-7B-Instruct",
+      description: "Examples: Qwen/Qwen2.5-72B-Instruct, deepseek-ai/DeepSeek-V3, meta-llama/Meta-Llama-3.1-8B-Instruct.",
+      visible: "${data.provider === 'siliconflow'}"
+    },
+    {
+      type: "group",
+      id: "openrouter",
+      label: "OpenRouter",
+      required: false,
+      visible: "${data.provider === 'openrouter'}",
+      description: "Multi-provider router at openrouter.ai/api/v1. Base URL is auto-filled."
+    },
+    {
+      type: "password",
+      key: "openrouter_api_key",
+      label: "OpenRouter API key",
+      required: true,
+      encrypted: true,
+      description: "sk-or-...",
+      visible: "${data.provider === 'openrouter'}"
+    },
+    {
+      type: "text",
+      key: "openrouter_model",
+      label: "Model",
+      required: false,
+      default: "openai/gpt-4o-mini",
+      description: "Format: provider/model (e.g. anthropic/claude-3.5-sonnet, deepseek/deepseek-chat).",
+      visible: "${data.provider === 'openrouter'}"
+    },
     // ── Generation defaults ──────────────────────────────────────
     {
       type: "group",
@@ -1648,6 +1811,20 @@ var aiTestActionHandler = async ({ values, payload }) => {
       message: `Vercel AI Gateway configured (model=${values.gateway_model}). Mount @objectstack/service-ai to exercise live calls.`
     };
   }
+  if (provider === "cloudflare") {
+    if (!values.cloudflare_account_id) {
+      return { ok: false, severity: "error", message: "Cloudflare account id is required." };
+    }
+    if (!values.cloudflare_api_key) {
+      return { ok: false, severity: "error", message: "Cloudflare AI Gateway token is required." };
+    }
+    const model2 = values.cloudflare_model ?? "(default openai/gpt-4o-mini)";
+    return {
+      ok: true,
+      severity: "info",
+      message: `Cloudflare AI Gateway configured (model=${model2}). Mount @objectstack/service-ai to exercise live calls.`
+    };
+  }
   const keyField = `${provider}_api_key`;
   if (!values[keyField]) {
     return { ok: false, severity: "error", message: `${provider} API key is required.` };
@@ -1677,8 +1854,8 @@ var manifest4 = {
   icon: "BookOpen",
   description: "Vector-store backend for RAG / knowledge sources. \u26A0 Switching adapter does NOT migrate existing indices \u2014 documents indexed under the previous adapter become unreachable until re-indexed. The embedder is configured separately under AI.",
   scope: "global",
-  readPermission: "setup.access",
-  writePermission: "setup.write",
+  readPermission: "manage_platform_settings",
+  writePermission: "manage_platform_settings",
   category: "Infrastructure",
   order: 35,
   specifiers: [