@billjr99/pi-openai-compat 1.0.2 → 1.1.9

package/README.md

@@ -40,9 +40,32 @@ If pi is already running when you install, type `/reload` first.
 | **OpenRouter** | `https://openrouter.ai/api/v1` | `sk-or-...` from openrouter.ai/keys |
 | **NVIDIA NIM** | `https://integrate.api.nvidia.com/v1` | `nvapi-...` from build.nvidia.com |
 | **Nous Research Portal** | `https://inference-api.nousresearch.com/v1` | Nous Portal API key |
+| **DeepSeek** | `https://api.deepseek.com/v1` | API key from platform.deepseek.com |
+| **xAI (Grok)** | `https://api.x.ai/v1` | API key from console.x.ai |
+| **Hugging Face** | `https://router.huggingface.co/v1` | `hf_...` token from huggingface.co/settings/tokens |
+| **Moonshot (Kimi)** | `https://api.moonshot.ai/v1` | `sk-...` key from platform.moonshot.ai |
+| **MiniMax** | `https://api.minimax.io/v1` | API key from platform.minimax.io |
+| **Z.ai** | `https://api.z.ai/api/paas/v4` | API key from z.ai |
+| **Venice AI** | `https://api.venice.ai/api/v1` | API key from venice.ai/settings/api |
+| **Cloudflare Workers AI** | `https://api.cloudflare.com/client/v4/accounts/{account}/ai/v1` | API token from dash.cloudflare.com |
+| **Cloudflare AI Gateway** | `https://gateway.ai.cloudflare.com/v1/{account}/{gateway}/openai` | API token from dash.cloudflare.com |
+| **Vercel AI Gateway** | `https://ai-gateway.vercel.sh/v1` | API key from vercel.com |
+| **OpenCode Zen** | `https://opencode.ai/zen/v1` | API key from opencode.ai |
 | **Ollama (local)** | `http://localhost:11434/v1` | Keyless |
+| **Ollama Cloud** | `https://ollama.com/v1` | Ollama Cloud API key from ollama.com |
 | **Custom** | Any URL you supply | Optional bearer token |
 
+> **Providers with built-in fallback model lists (as of May 2026)**
+> Some providers do not support `GET /v1/models` or return a non-JSON response.
+> For these, the extension substitutes a built-in list and shows a warning during
+> `/compat-login`.
+>
+> | Provider | Reason |
+> |---|---|
+> | **Cloudflare Workers AI** | Returns HTTP 405 — method not supported |
+> | **Cloudflare AI Gateway** | Returns HTTP 401 — no anonymous model enumeration |
+> | **Hugging Face** | Returns HTML rather than JSON for `/v1/models` |
+
 ---
 
 ## Commands
@@ -171,6 +194,65 @@ npm publish --access public
 
 `npm version` also creates a git tag, so GitHub gets release tags automatically.
 
+### Automated publishing via GitHub Actions
+
+Every push to `main` automatically bumps the patch version and publishes to npm
+via `.github/workflows/publish.yml`. This workflow uses **npm trusted publishing**
+(OIDC) — no long-lived token or secret is needed anywhere.
+
+#### Required one-time setup on npmjs.com
+
+> **Do this before the first automated publish or it will fail.**
+
+1. Go to **npmjs.com** → sign in → click your avatar → **Packages** →
+   `@billjr99/pi-openai-compat`
+2. Click **Settings** (in the left sidebar of the package page)
+3. Scroll to **Publishing** → enable **Trusted Publishing**
+4. Add a publisher with these exact values:
+
+| Field | Value |
+|---|---|
+| Repository owner | `BillJr99` |
+| Repository name | `pi-openai-compat` |
+| Workflow filename | `publish.yml` |
+| Environment | *(leave blank)* |
+
+5. Click **Save**
+
+No token to copy, no secret to rotate. That's the only setup step.
+
+#### What the workflow does
+
+```yaml
+# .github/workflows/publish.yml
+on:
+  push:
+    branches: [main]
+
+permissions:
+  contents: write    # push the version-bump commit back to main
+  id-token: write    # request an OIDC token from GitHub for npm auth
+```
+
+On every push to `main` the workflow:
+
+1. Checks out the repo with `GITHUB_TOKEN` so it can push back
+2. Runs `npm version patch --no-git-tag-version` to increment the patch number
+   in `package.json` (e.g. `1.1.0` → `1.1.1`)
+3. Commits the updated `package.json` with `[skip ci]` in the message so the
+   commit does not re-trigger the workflow
+4. Pushes the bump commit to `main`
+5. Runs `npm publish --access public --provenance`
+   - GitHub mints a short-lived OIDC token proving the job is running from
+     this exact repo and workflow
+   - npm validates it against the trusted publisher config above and allows
+     the publish — no `NPM_TOKEN` secret required
+   - `--provenance` attaches a signed build attestation to the package,
+     visible on the npmjs.com package page under **Provenance**
+
+The published package is immediately available at
+https://www.npmjs.com/package/@billjr99/pi-openai-compat.
+
 ### Installing from npm
 
 ```bash

package/index.ts

@@ -60,111 +60,155 @@ const TEMPLATES: Record<string, {
   modelFilter?: string[];
   /** If true, prompt the user to confirm/edit the base URL (like ollama/custom). */
   promptUrl?: boolean;
+  /** Used when the provider does not support GET /v1/models (e.g. returns 405). */
+  fallbackModels?: string[];
+  /** Where to obtain the API key; presence implies the key is required. */
+  keyHint?: string;
 }> = {
   openrouter: {
     displayName: "OpenRouter",
     baseUrl: "https://openrouter.ai/api/v1",
     keyless: false,
+    keyHint: "openrouter.ai/keys",
   },
   nvidia_nim: {
     displayName: "NVIDIA NIM",
     baseUrl: "https://integrate.api.nvidia.com/v1",
     keyless: false,
+    keyHint: "build.nvidia.com",
   },
   nous: {
     displayName: "Nous Research Portal",
     baseUrl: "https://inference-api.nousresearch.com/v1",
     keyless: false,
+    keyHint: "nousresearch.com",
   },
   google: {
     displayName: "Google Gemini",
     baseUrl: "https://generativelanguage.googleapis.com/v1beta/openai",
     keyless: false,
-    modelFilter: [
-      "gemini-2.5-flash",
-      "gemini-2.5-flash-lite",
-      "gemini-3-flash-preview",
-      "gemini-3.1-flash-lite-preview",
-      "gemini-3.1-pro-preview",
-    ],
+    keyHint: "aistudio.google.com/apikey",
   },
   cerebras: {
     displayName: "Cerebras",
     baseUrl: "https://api.cerebras.ai/v1",
     keyless: false,
-    modelFilter: ["qwen3-235b"],
+    keyHint: "cloud.cerebras.ai",
   },
   github_models: {
     displayName: "GitHub Models",
     baseUrl: "https://models.inference.ai.azure.com",
     keyless: false,
-    modelFilter: ["gpt-4o", "openai/gpt-4.1"],
+    keyHint: "github.com/settings/tokens",
   },
   sambanova: {
     displayName: "SambaNova",
     baseUrl: "https://api.sambanova.ai/v1",
     keyless: false,
-    modelFilter: [
-      "Meta-Llama-3.3-70B-Instruct",
-      "DeepSeek-V3.1",
-      "DeepSeek-V3.2",
-      "gpt-oss-120b",
-      "Llama-4-Maverick-17B-128E-Instruct",
-      "gemma-3-12b-it",
-    ],
+    keyHint: "cloud.sambanova.ai",
   },
   mistral: {
     displayName: "Mistral",
     baseUrl: "https://api.mistral.ai/v1",
     keyless: false,
-    modelFilter: [
-      "mistral-large-latest",
-      "mistral-medium-latest",
-      "magistral-medium-latest",
-      "codestral-latest",
-      "devstral-latest",
-    ],
+    keyHint: "console.mistral.ai/api-keys",
   },
   groq: {
     displayName: "Groq",
     baseUrl: "https://api.groq.com/openai/v1",
     keyless: false,
-    modelFilter: [
-      "llama-3.3-70b-versatile",
-      "meta-llama/llama-4-scout-17b-16e-instruct",
-      "openai/gpt-oss-120b",
-      "openai/gpt-oss-20b",
-      "qwen/qwen3-32b",
-      "llama-3.1-8b-instant",
-    ],
+    keyHint: "console.groq.com/keys",
   },
-  cloudflare: {
+  cloudflare_workers: {
     displayName: "Cloudflare Workers AI",
     baseUrl: "https://api.cloudflare.com/client/v4/accounts/YOUR_ACCOUNT_ID/ai/v1",
     keyless: false,
     promptUrl: true,
-    modelFilter: [
+    keyHint: "dash.cloudflare.com → My Profile → API Tokens",
+    // Cloudflare Workers AI returns 405 for GET /v1/models; use a curated list.
+    fallbackModels: [
+      "@cf/meta/llama-4-scout-17b-16e-instruct",
       "@cf/meta/llama-3.3-70b-instruct-fp8-fast",
+      "@cf/meta/llama-3.1-8b-instruct",
+      "@cf/qwen/qwen3-30b-a3b-fp8",
+      "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b",
+    ],
+  },
+  cloudflare_ai_gateway: {
+    displayName: "Cloudflare AI Gateway",
+    // YOUR_PROVIDER is the upstream slug (e.g. "workers-ai", "openai"). /v1 is
+    // appended so that fetchModels and chat completions hit the correct path.
+    baseUrl: "https://gateway.ai.cloudflare.com/v1/YOUR_ACCOUNT_ID/YOUR_GATEWAY_SLUG/YOUR_PROVIDER/v1",
+    keyless: false,
+    promptUrl: true,
+    keyHint: "dash.cloudflare.com → My Profile → API Tokens",
+    fallbackModels: [
       "@cf/meta/llama-4-scout-17b-16e-instruct",
-      "@cf/openai/gpt-oss-120b",
-      "@cf/zai-org/glm-4.7-flash",
-      "@cf/moonshotai/kimi-k2.5",
-      "@cf/moonshotai/kimi-k2.6",
+      "@cf/meta/llama-3.3-70b-instruct-fp8-fast",
+      "@cf/meta/llama-3.1-8b-instruct",
       "@cf/qwen/qwen3-30b-a3b-fp8",
       "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b",
-      "@cf/ibm-granite/granite-4.0-h-micro",
     ],
   },
   zhipu: {
     displayName: "Zhipu (Z.ai / BigModel)",
     baseUrl: "https://open.bigmodel.cn/api/paas/v4",
     keyless: false,
-    modelFilter: ["glm-4.5-flash", "glm-4.7-flash"],
+    keyHint: "open.bigmodel.cn/usercenter/apikeys",
+  },
+  zai: {
+    displayName: "Z.ai",
+    baseUrl: "https://api.z.ai/api/paas/v4",
+    keyless: false,
+    keyHint: "platform.z.ai",
   },
   cohere: {
     displayName: "Cohere",
     baseUrl: "https://api.cohere.com/compatibility/v1",
     keyless: false,
+    keyHint: "dashboard.cohere.com/api-keys",
+  },
+  deepseek: {
+    displayName: "DeepSeek",
+    baseUrl: "https://api.deepseek.com/v1",
+    keyless: false,
+    keyHint: "platform.deepseek.com/api_keys",
+  },
+  xai: {
+    displayName: "xAI (Grok)",
+    baseUrl: "https://api.x.ai/v1",
+    keyless: false,
+    keyHint: "console.x.ai",
+  },
+  huggingface: {
+    displayName: "Hugging Face",
+    baseUrl: "https://router.huggingface.co/v1",
+    keyless: false,
+    keyHint: "huggingface.co/settings/tokens",
+    fallbackModels: [
+      "meta-llama/Llama-3.3-70B-Instruct",
+      "meta-llama/Meta-Llama-3-8B-Instruct",
+      "mistralai/Mistral-7B-Instruct-v0.3",
+      "Qwen/Qwen2.5-72B-Instruct",
+    ],
+  },
+  moonshot: {
+    displayName: "Moonshot (Kimi)",
+    baseUrl: "https://api.moonshot.ai/v1",
+    keyless: false,
+    keyHint: "platform.moonshot.ai/console/api-key",
+  },
+  minimax: {
+    displayName: "MiniMax",
+    baseUrl: "https://api.minimax.io/v1",
+    keyless: false,
+    keyHint: "platform.minimax.io",
+  },
+  venice: {
+    displayName: "Venice AI",
+    baseUrl: "https://api.venice.ai/api/v1",
+    keyless: false,
+    keyHint: "venice.ai/settings/api",
   },
   ollama: {
     displayName: "Ollama (local, keyless)",
@@ -172,6 +216,32 @@ const TEMPLATES: Record<string, {
     keyless: true,
     promptUrl: true,
   },
+  ollama_cloud: {
+    displayName: "Ollama Cloud",
+    baseUrl: "https://ollama.com/v1",
+    keyless: false,
+    keyHint: "ollama.com/settings/api-keys",
+  },
+  llmproxy: {
+    displayName: "llmproxy (local)",
+    baseUrl: "http://localhost:8080/v1",
+    keyless: true,
+    promptUrl: true,
+  },
+  vercel: {
+    displayName: "Vercel AI Gateway",
+    baseUrl: "https://ai-gateway.vercel.sh/v1",
+    keyless: false,
+    keyHint: "vercel.com/account/tokens",
+  },
+  opencode_zen: {
+    displayName: "OpenCode Zen",
+    baseUrl: "https://opencode.ai/zen/v1",
+    keyless: false,
+    keyHint: "opencode.ai",
+    // Verify current model IDs via GET /v1/models — the free model list may change.
+    modelFilter: ["big-pickle", "deepseek-v4-flash-free", "minimax-m2.5-free", "nemotron-3-super-free"],
+  },
   custom: {
     displayName: "Custom Endpoint",
     baseUrl: "",
@@ -226,6 +296,21 @@ function saveConfig(config: ExtensionConfig): void {
 // Networking
 // ─────────────────────────────────────────────────────────────────────────────
 
+function isLocalUrl(url: string): boolean {
+  try {
+    const { hostname } = new URL(url);
+    return (
+      hostname === "localhost" ||
+      hostname === "127.0.0.1" ||
+      hostname === "::1" ||
+      hostname.endsWith(".local") ||
+      hostname.endsWith(".localhost")
+    );
+  } catch {
+    return false;
+  }
+}
+
 async function fetchModels(baseUrl: string, apiKey: string | null): Promise<CachedModel[]> {
   const url = `${baseUrl.replace(/\/+$/, "")}/models`;
   const headers: Record<string, string> = { Accept: "application/json" };
@@ -268,9 +353,9 @@ function compatKey(key: string): string {
 
 function registerProvider(pi: ExtensionAPI, key: string, p: ProviderConfig): void {
   pi.registerProvider(compatKey(key), {
-    name: p.displayName,
+    name: `compat/${key.replace(/_/g, "-")}`,
     baseUrl: p.baseUrl,
-    apiKey: p.apiKey ?? "ollama",
+    apiKey: p.apiKey ?? (isLocalUrl(p.baseUrl) ? "local" : ""),
     api: "openai-completions" as const,
     models: buildProviderModels(p.cachedModels),
   });
@@ -359,8 +444,7 @@ export default async function (pi: ExtensionAPI) {
 
       // Step 2 — base URL
       let baseUrl = tpl.baseUrl;
-      if (key === "cloudflare") {
-        // Ask for just the account ID and splice it into the template URL.
+      if (key === "cloudflare_workers") {
         const entered = await ctx.ui.input(
           "Account ID",
           "Your Cloudflare Account ID (find it on the Cloudflare dashboard overview page):",
@@ -370,10 +454,41 @@ export default async function (pi: ExtensionAPI) {
         const accountId = entered.trim();
         if (!accountId) { ctx.ui.notify("Account ID cannot be empty.", "error"); return; }
         baseUrl = tpl.baseUrl.replace("YOUR_ACCOUNT_ID", accountId);
+      } else if (key === "cloudflare_ai_gateway") {
+        const accountIdInput = await ctx.ui.input(
+          "Account ID",
+          "Your Cloudflare Account ID (find it on the Cloudflare dashboard overview page):",
+          ""
+        );
+        if (accountIdInput == null) { ctx.ui.notify("Login cancelled.", "info"); return; }
+        const accountId = accountIdInput.trim();
+        if (!accountId) { ctx.ui.notify("Account ID cannot be empty.", "error"); return; }
+
+        const gatewayInput = await ctx.ui.input(
+          "Gateway Name",
+          "Your AI Gateway name/slug (find it under AI → AI Gateway in the Cloudflare dashboard):",
+          ""
+        );
+        if (gatewayInput == null) { ctx.ui.notify("Login cancelled.", "info"); return; }
+        const gatewaySlug = gatewayInput.trim();
+        if (!gatewaySlug) { ctx.ui.notify("Gateway name cannot be empty.", "error"); return; }
+
+        const providerInput = await ctx.ui.input(
+          "Provider",
+          "Upstream provider slug (e.g. openai, workers-ai, anthropic — must match your gateway config):",
+          "openai"
+        );
+        if (providerInput == null) { ctx.ui.notify("Login cancelled.", "info"); return; }
+        const provider = providerInput.trim() || "openai";
+
+        baseUrl = tpl.baseUrl
+          .replace("YOUR_ACCOUNT_ID", accountId)
+          .replace("YOUR_GATEWAY_SLUG", gatewaySlug)
+          .replace("YOUR_PROVIDER", provider);
       } else if (tpl.promptUrl) {
         const defaultUrl = tpl.baseUrl;
-        const prompt = key === "ollama"
-          ? `Ollama base URL — press Enter for default (${defaultUrl}):`
+        const prompt = isLocalUrl(defaultUrl)
+          ? `Base URL — press Enter for default (${defaultUrl}):`
           : "Base URL of your endpoint (e.g. https://api.example.com/v1):";
         const entered = await ctx.ui.input("Base URL", prompt, defaultUrl);
         if (entered == null) { ctx.ui.notify("Login cancelled.", "info"); return; }
@@ -381,14 +496,13 @@ export default async function (pi: ExtensionAPI) {
         if (!baseUrl) { ctx.ui.notify("Base URL cannot be empty.", "error"); return; }
       }
 
-      // Step 3 — API key
+      // Step 3 — API key (skipped for keyless templates and detected local URLs)
       let apiKey: string | null = null;
-      if (!tpl.keyless) {
-        const entered = await ctx.ui.input(
-          "API Key",
-          "Your API key — leave blank if keyless:",
-          ""
-        );
+      if (!tpl.keyless && !isLocalUrl(baseUrl)) {
+        const keyPrompt = tpl.keyHint
+          ? `Your API key (required) — get it at ${tpl.keyHint}:`
+          : "Your API key — leave blank if keyless:";
+        const entered = await ctx.ui.input("API Key", keyPrompt, "");
         if (entered == null) { ctx.ui.notify("Login cancelled.", "info"); return; }
         apiKey = entered.trim() || null;
       }
@@ -399,8 +513,16 @@ export default async function (pi: ExtensionAPI) {
       try {
         models = await fetchModels(baseUrl, apiKey);
       } catch (err) {
-        ctx.ui.notify(`Connection failed — not saved.\n${err}`, "error");
-        return;
+        if (tpl.fallbackModels && tpl.fallbackModels.length > 0) {
+          ctx.ui.notify(
+            `Could not fetch model list from ${tpl.displayName} (${err}).\nUsing built-in model list instead.`,
+            "warning"
+          );
+          models = tpl.fallbackModels.map((id) => ({ id }));
+        } else {
+          ctx.ui.notify(`Connection failed — not saved.\n${err}`, "error");
+          return;
+        }
       }
       if (!models.length) {
         ctx.ui.notify("Connected but no models returned. Check URL and key.", "error");

package/package.json

@@ -1,13 +1,13 @@
 {
   "name": "@billjr99/pi-openai-compat",
-  "version": "1.0.2",
+  "version": "1.1.9",
   "description": "pi-coding-agent extension: OpenAI-compatible endpoint support (OpenRouter, NVIDIA NIM, Nous Portal, Ollama, custom)",
   "author": "Bill Mongan <https://github.com/BillJr99>",
   "license": "MIT",
   "homepage": "https://github.com/BillJr99/pi-openai-compat#readme",
   "repository": {
     "type": "git",
-    "url": "https://github.com/BillJr99/pi-openai-compat.git"
+    "url": "git+https://github.com/BillJr99/pi-openai-compat.git"
   },
   "bugs": {
     "url": "https://github.com/BillJr99/pi-openai-compat/issues"
@@ -17,7 +17,11 @@
     "README.md"
   ],
   "pi": {
-    "extensions": ["./index.ts"]
+    "extensions": [
+      "./index.ts"
+    ]
   },
-  "dependencies": {}
+  "devDependencies": {
+    "typescript": "^6.0.3"
+  }
 }