@elizaos/plugin-openai 2.0.0-alpha.9 → 2.0.3-beta.2

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Shaw Walters and elizaOS Contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,221 @@
+# @elizaos/plugin-openai
+
+OpenAI model-provider plugin for [elizaOS](https://github.com/elizaos/eliza). Adds text generation, embeddings, image generation, image description, audio transcription, text-to-speech, and deep research to any Eliza agent by registering handlers for the standard `ModelType.*` slots on the `AgentRuntime`.
+
+## What this plugin does
+
+- **Text generation** — multiple tiers (nano, small, medium, large, mega), plus dedicated response-handler and action-planner slots. Supports streaming and structured JSON output.
+- **Text embeddings** — `text-embedding-3-small` by default; dimension configurable.
+- **Image generation** — DALL-E 3 by default (`dall-e-3`).
+- **Image description** — vision model analyzes an image URL and returns `{ title, description }`.
+- **Audio transcription** — speech-to-text (`gpt-5-mini-transcribe` by default); accepts `Buffer`, `Blob`, `File`, or a URL.
+- **Text-to-speech** — returns an `ArrayBuffer` of audio. Six voices; mp3/wav/flac/opus/aac/pcm output.
+- **Deep research** — `ModelType.RESEARCH` via the OpenAI Responses API (`o3-deep-research` by default); returns annotated, multi-source research reports.
+- **Tokenizer** — encode/decode using js-tiktoken (browser-safe, no network calls).
+
+Works with any OpenAI-compatible endpoint: OpenAI, Cerebras, EvoLink, OpenRouter, local servers, etc.
+
+## Enabling the plugin
+
+Add `@elizaos/plugin-openai` to your character's plugin list:
+
+```json
+{
+  "plugins": ["@elizaos/plugin-openai"]
+}
+```
+
+The plugin auto-enables when `OPENAI_API_KEY`, `CEREBRAS_API_KEY`, or `EVOLINK_API_KEY` is present in the environment.
+
+## Required configuration
+
+```
+OPENAI_API_KEY=sk-...
+```
+
+That is the only required setting. All other settings are optional overrides.
+
+## Full configuration reference
+
+Set these as environment variables or in your character's `settings` object.
+
+### API endpoint
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_BASE_URL` | `https://api.openai.com/v1` | Override for compatible endpoints |
+| `ELIZA_PROVIDER` | — | Set to `cerebras` or `evolink` to force a compatible-provider mode |
+| `EVOLINK_API_KEY` | — | Auth for EvoLink's OpenAI-compatible endpoint |
+| `EVOLINK_BASE_URL` | `https://direct.evolink.ai/v1` | EvoLink endpoint used when `EVOLINK_API_KEY` enables the plugin |
+| `EVOLINK_MODEL` | `gpt-5.2` | Text model used by EvoLink mode |
+
+### Text generation models
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_SMALL_MODEL` / `SMALL_MODEL` | `gpt-5.4-mini` | Fast, cost-effective responses |
+| `OPENAI_NANO_MODEL` / `NANO_MODEL` | falls back to small | Fastest tier |
+| `OPENAI_MEDIUM_MODEL` / `MEDIUM_MODEL` | falls back to small | Mid-tier |
+| `OPENAI_LARGE_MODEL` / `LARGE_MODEL` | `gpt-5` | Complex reasoning |
+| `OPENAI_MEGA_MODEL` / `MEGA_MODEL` | falls back to large | Highest tier |
+| `OPENAI_RESPONSE_HANDLER_MODEL` | falls back to small | Response-handler slot |
+| `OPENAI_ACTION_PLANNER_MODEL` | falls back to medium | Action-planner slot |
+| `OPENAI_REASONING_EFFORT` | — | `minimal`/`low`/`medium`/`high` for o-series models |
+
+### Embeddings
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_EMBEDDING_MODEL` | `text-embedding-3-small` | Embedding model |
+| `OPENAI_EMBEDDING_URL` | `OPENAI_BASE_URL` | Separate endpoint for embeddings |
+| `OPENAI_EMBEDDING_API_KEY` | `OPENAI_API_KEY` | Separate key for embeddings |
+| `OPENAI_EMBEDDING_DIMENSIONS` | `1536` | Vector dimensions (must match model) |
+
+### Image generation and description
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_IMAGE_MODEL` | `dall-e-3` | Image generation model |
+| `OPENAI_IMAGE_DESCRIPTION_MODEL` | `gpt-5-mini` | Vision/description model |
+| `OPENAI_IMAGE_DESCRIPTION_BASE_URL` | `OPENAI_BASE_URL` | Separate endpoint for vision |
+| `OPENAI_IMAGE_DESCRIPTION_API_KEY` | `OPENAI_API_KEY` | Separate key for vision |
+| `OPENAI_IMAGE_DESCRIPTION_MAX_TOKENS` | `8192` | Max tokens in description response |
+
+### Audio
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_TRANSCRIPTION_MODEL` | `gpt-5-mini-transcribe` | Speech-to-text model |
+| `OPENAI_TTS_MODEL` | `tts-1` | Text-to-speech model |
+| `OPENAI_TTS_VOICE` | `nova` | Voice: alloy, echo, fable, onyx, nova, shimmer |
+| `OPENAI_TTS_INSTRUCTIONS` | — | Style instructions for TTS |
+
+### Deep research
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_RESEARCH_MODEL` | `o3-deep-research` | Research model (o3 or o4-mini variants) |
+| `OPENAI_RESEARCH_TIMEOUT` | `3600000` (1 hr) | Request timeout in milliseconds |
+
+### Browser and proxy
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_BROWSER_BASE_URL` | — | Proxy URL for browser builds (keeps key server-side) |
+| `OPENAI_BROWSER_EMBEDDING_URL` | — | Proxy URL for browser embedding requests |
+| `OPENAI_ALLOW_BROWSER_API_KEY` | `false` | Send auth header in browser builds (opt-in) |
+
+### Other
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENAI_EXPERIMENTAL_TELEMETRY` | `false` | Enable AI SDK experimental telemetry |
+
+## Usage examples
+
+```ts
+import { ModelType } from "@elizaos/core";
+
+// Text generation
+const reply = await runtime.useModel(ModelType.TEXT_LARGE, {
+  prompt: "Explain quantum entanglement in plain English.",
+});
+
+// Streaming text
+const result = await runtime.useModel(ModelType.TEXT_LARGE, {
+  prompt: "Count from 1 to 10.",
+  stream: true,
+  onStreamChunk: (chunk) => process.stdout.write(chunk),
+});
+
+// Structured JSON output
+const data = await runtime.useModel(ModelType.TEXT_LARGE, {
+  prompt: "Return a JSON object with name and age fields.",
+  responseSchema: {
+    type: "object",
+    properties: { name: { type: "string" }, age: { type: "number" } },
+    required: ["name", "age"],
+  },
+});
+
+// Embedding
+const vector = await runtime.useModel(ModelType.TEXT_EMBEDDING, {
+  text: "text to embed",
+});
+
+// Image generation
+const images = await runtime.useModel(ModelType.IMAGE, {
+  prompt: "A sunset over mountains",
+  count: 1,
+  size: "1024x1024",
+});
+
+// Image description
+const { title, description } = await runtime.useModel(
+  ModelType.IMAGE_DESCRIPTION,
+  "https://example.com/photo.jpg"
+);
+
+// Audio transcription (Buffer, Blob, File, or URL string all accepted)
+const transcript = await runtime.useModel(ModelType.TRANSCRIPTION, audioBuffer);
+
+// Text-to-speech
+const audio = await runtime.useModel(ModelType.TEXT_TO_SPEECH, {
+  text: "Hello, world.",
+  voice: "nova",
+  format: "mp3",
+});
+
+// Deep research (may take minutes)
+const report = await runtime.useModel(ModelType.RESEARCH, {
+  input: "What are the latest advances in fusion energy?",
+  tools: [{ type: "web_search_preview" }],
+});
+console.log(report.text, report.annotations);
+```
+
+## Browser proxy setup
+
+In browser builds this plugin does not send `Authorization` headers by default, to avoid exposing API keys in frontend bundles. Point `OPENAI_BROWSER_BASE_URL` at a server-side proxy that injects the key:
+
+```ts
+// Minimal Express proxy
+import express from "express";
+const app = express();
+app.use(express.json());
+
+app.all("/openai/*", async (req, res) => {
+  const url = `https://api.openai.com/v1/${req.params[0]}`;
+  const r = await fetch(url, {
+    method: req.method,
+    headers: { Authorization: `Bearer ${process.env.OPENAI_API_KEY}`, "Content-Type": "application/json" },
+    body: req.method !== "GET" ? JSON.stringify(req.body) : undefined,
+  });
+  res.status(r.status).send(await r.text());
+});
+
+app.listen(3000);
+```
+
+Then set `OPENAI_BROWSER_BASE_URL=http://localhost:3000/openai`.
+
+## Cerebras compatibility
+
+Point `OPENAI_BASE_URL` at a Cerebras endpoint or set `ELIZA_PROVIDER=cerebras` and the plugin automatically adapts: structured output uses `json_object` mode, `reasoning_effort` defaults to `"low"` for reasoning-capable models (to prevent empty responses), and `CEREBRAS_API_KEY` is accepted as an alias for `OPENAI_API_KEY`. Embeddings fall back to a deterministic local hash when no explicit embedding URL is set, since Cerebras does not provide an embeddings endpoint.
+
+## EvoLink compatibility
+
+Set `EVOLINK_API_KEY` to use EvoLink through its OpenAI-compatible endpoint. The plugin defaults to `https://direct.evolink.ai/v1` and `gpt-5.2`; set `EVOLINK_BASE_URL` or `EVOLINK_MODEL` to override either value.
+
+## Prompt caching
+
+Pass `providerOptions.openai.promptCacheKey` and `promptCacheRetention` on any `GenerateTextParams` call to enable OpenAI prompt caching:
+
+```ts
+await runtime.useModel(ModelType.TEXT_LARGE, {
+  prompt: "...",
+  providerOptions: {
+    openai: { promptCacheKey: "my-key", promptCacheRetention: "24h" },
+  },
+});
+```

package/auto-enable.ts

@@ -0,0 +1,17 @@
+// Auto-enable check for @elizaos/plugin-openai.
+//
+// Plugin manifest entry-point — referenced by package.json's
+// `elizaos.plugin.autoEnableModule`. Keep this module light: env reads only,
+// no service init, no transitive imports of the full plugin runtime. The
+// auto-enable engine loads dozens of these per boot.
+import type { PluginAutoEnableContext } from "@elizaos/core";
+
+const ENV_KEYS = ["OPENAI_API_KEY", "CEREBRAS_API_KEY", "EVOLINK_API_KEY"] as const;
+
+/** Enable when an OpenAI-compatible API key is present in the environment. */
+export function shouldEnable(ctx: PluginAutoEnableContext): boolean {
+  return ENV_KEYS.some((k) => {
+    const v = ctx.env[k];
+    return typeof v === "string" && v.trim() !== "";
+  });
+}

package/package.json

@@ -1,8 +1,8 @@
 {
   "name": "@elizaos/plugin-openai",
-  "version": "2.0.0-alpha.9",
+  "version": "2.0.3-beta.2",
   "type": "module",
-  "main": "dist/cjs/index.node.cjs",
+  "main": "dist/node/index.node.js",
   "module": "dist/node/index.node.js",
   "types": "dist/index.d.ts",
   "browser": "dist/browser/index.browser.js",
@@ -23,44 +23,65 @@
       "node": {
         "types": "./dist/node/index.d.ts",
         "import": "./dist/node/index.node.js",
-        "require": "./dist/cjs/index.node.cjs",
         "default": "./dist/node/index.node.js"
       },
       "default": "./dist/node/index.node.js"
+    },
+    "./*.css": "./dist/*.css",
+    "./*": {
+      "types": "./dist/*.d.ts",
+      "import": "./dist/*.js",
+      "default": "./dist/*.js"
     }
   },
   "files": [
     "dist",
-    "rust"
+    "rust",
+    "auto-enable.ts"
   ],
+  "elizaos": {
+    "plugin": {
+      "autoEnableModule": "./auto-enable.ts",
+      "capabilities": [
+        "text-large",
+        "text-small",
+        "tool-use",
+        "embedding",
+        "text-to-speech",
+        "image-description"
+      ]
+    }
+  },
   "dependencies": {
     "@ai-sdk/openai": "^3.0.9",
-    "@elizaos/core": "alpha",
     "ai": "^6.0.30",
-    "js-tiktoken": "^1.0.21",
-    "undici": "^7.16.0"
+    "js-tiktoken": "^1.0.21"
   },
   "devDependencies": {
-    "@biomejs/biome": "^2.3.11",
+    "@biomejs/biome": "^2.4.14",
+    "@elizaos/core": "2.0.3-beta.2",
     "@types/bun": "^1.3.8",
     "@types/json-schema": "^7.0.15",
     "@types/node": "^25.0.3",
-    "typescript": "^5.9.3"
+    "typescript": "^6.0.3",
+    "uuid": "^14.0.0",
+    "vitest": "^4.0.0"
   },
   "peerDependencies": {
-    "zod": "^4.3.6"
+    "@elizaos/core": "2.0.3-beta.2",
+    "zod": "^4.4.3"
   },
   "scripts": {
     "build:ts": "bun run build.ts",
     "dev": "bun --hot build.ts",
     "lint": "bunx @biomejs/biome check --write --unsafe .",
     "lint:check": "bunx @biomejs/biome check .",
-    "clean": "rm -rf dist   .turbo node_modules",
+    "clean": "rm -rf dist .turbo",
     "format": "bunx @biomejs/biome format --write .",
     "format:check": "bunx @biomejs/biome format .",
-    "typecheck": "tsc --noEmit",
-    "test": "vitest run __tests__/ --passWithNoTests",
-    "test:ts": "vitest run __tests__/ --passWithNoTests",
+    "typecheck": "tsgo --noEmit",
+    "test": "vitest run --config ./vitest.config.ts",
+    "test:ts": "vitest run --config ./vitest.config.ts",
     "build": "bun run build.ts"
   },
   "publishConfig": {
@@ -82,6 +103,26 @@
         "default": "https://api.openai.com/v1",
         "sensitive": false
       },
+      "EVOLINK_API_KEY": {
+        "type": "string",
+        "description": "API key used to authenticate requests to EvoLink's OpenAI-compatible API.",
+        "required": false,
+        "sensitive": true
+      },
+      "EVOLINK_BASE_URL": {
+        "type": "string",
+        "description": "Base URL for EvoLink's OpenAI-compatible API.",
+        "required": false,
+        "default": "https://direct.evolink.ai/v1",
+        "sensitive": false
+      },
+      "EVOLINK_MODEL": {
+        "type": "string",
+        "description": "Identifier of the EvoLink chat model to use when EVOLINK_API_KEY enables this plugin.",
+        "required": false,
+        "default": "gpt-5.2",
+        "sensitive": false
+      },
       "OPENAI_SMALL_MODEL": {
         "type": "string",
         "description": "Identifier of the small language model to be used (overrides SMALL_MODEL).",
@@ -92,7 +133,7 @@
         "type": "string",
         "description": "Fallback identifier for the small language model if OPENAI_SMALL_MODEL is not set.",
         "required": false,
-        "default": "gpt-5-mini",
+        "default": "gpt-5.4-mini",
         "sensitive": false
       },
       "OPENAI_LARGE_MODEL": {
@@ -134,6 +175,18 @@
         "default": 1536,
         "sensitive": false
       },
+      "OPENAI_IMAGE_DESCRIPTION_API_KEY": {
+        "type": "string",
+        "description": "Optional API key for image-description requests. Falls back to the general OpenAI API key when unset.",
+        "required": false,
+        "sensitive": true
+      },
+      "OPENAI_IMAGE_DESCRIPTION_BASE_URL": {
+        "type": "string",
+        "description": "Optional OpenAI-compatible base URL for image-description requests. Useful when text models use a non-vision-compatible endpoint.",
+        "required": false,
+        "sensitive": false
+      },
       "OPENAI_IMAGE_DESCRIPTION_MODEL": {
         "type": "string",
         "description": "Identifier of the model used for describing images.",
@@ -189,8 +242,8 @@
       }
     }
   },
-  "gitHead": "646c632924826e2b75c2304a75ee56959fe4a460",
-  "milady": {
+  "gitHead": "82fe0f44215954c2417328203f5bd6510985c1fc",
+  "eliza": {
     "platforms": [
       "browser",
       "node"

package/dist/browser/index.browser.js

@@ -1,3 +0,0 @@
-import{logger as l,ModelType as I}from"@elizaos/core";import{logger as _}from"@elizaos/core";import{logger as z}from"@elizaos/core";function xo(o){if(typeof process>"u"||!process.env)return;let n=process.env[o];return n===void 0?void 0:String(n)}function s(o,n,c){let r=o.getSetting(n);if(r!==void 0&&r!==null)return String(r);return xo(n)??c}function K(o,n,c){let r=s(o,n);if(r===void 0)return c;let f=Number.parseInt(r,10);if(!Number.isFinite(f))throw Error(`Setting '${n}' must be a valid integer, got: ${r}`);return f}function v(o,n,c){let r=s(o,n);if(r===void 0)return c;let f=r.toLowerCase();return f==="true"||f==="1"||f==="yes"}function E(){return typeof globalThis<"u"&&typeof globalThis.document<"u"}function g(o){return E()&&!!s(o,"OPENAI_BROWSER_BASE_URL")}function J(o){return s(o,"OPENAI_API_KEY")}function Ko(o){let n=s(o,"OPENAI_EMBEDDING_API_KEY");if(n)return z.debug("[OpenAI] Using specific embedding API key"),n;return z.debug("[OpenAI] Falling back to general API key for embeddings"),J(o)}function R(o,n=!1){if(E()&&!v(o,"OPENAI_ALLOW_BROWSER_API_KEY",!1))return{};let c=n?Ko(o):J(o);return c?{Authorization:`Bearer ${c}`}:{}}function O(o){let n=s(o,"OPENAI_BROWSER_BASE_URL"),c=E()&&n?n:s(o,"OPENAI_BASE_URL")??"https://api.openai.com/v1";return z.debug(`[OpenAI] Base URL: ${c}`),c}function u(o){let n=E()?s(o,"OPENAI_BROWSER_EMBEDDING_URL")??s(o,"OPENAI_BROWSER_BASE_URL"):s(o,"OPENAI_EMBEDDING_URL");if(n)return z.debug(`[OpenAI] Using embedding base URL: ${n}`),n;return z.debug("[OpenAI] Falling back to general base URL for embeddings"),O(o)}function F(o){return s(o,"OPENAI_SMALL_MODEL")??s(o,"SMALL_MODEL")??"gpt-5-mini"}function d(o){return s(o,"OPENAI_LARGE_MODEL")??s(o,"LARGE_MODEL")??"gpt-5"}function m(o){return s(o,"OPENAI_EMBEDDING_MODEL")??"text-embedding-3-small"}function oo(o){return s(o,"OPENAI_IMAGE_DESCRIPTION_MODEL")??"gpt-5-mini"}function no(o){return s(o,"OPENAI_TRANSCRIPTION_MODEL")??"gpt-5-mini-transcribe"}function co(o){return s(o,"OPENAI_TTS_MODEL")??"tts-1"}function ro(o){return s(o,"OPENAI_TTS_VOICE")??"nova"}function fo(o){return s(o,"OPENAI_TTS_INSTRUCTIONS")??""}function To(o){return s(o,"OPENAI_IMAGE_MODEL")??"dall-e-3"}function io(o){return v(o,"OPENAI_EXPERIMENTAL_TELEMETRY",!1)}function ko(o){return K(o,"OPENAI_EMBEDDING_DIMENSIONS",1536)}function Ao(o){return K(o,"OPENAI_IMAGE_DESCRIPTION_MAX_TOKENS",8192)}function to(o){return s(o,"OPENAI_RESEARCH_MODEL")??"o3-deep-research"}function Po(o){return K(o,"OPENAI_RESEARCH_TIMEOUT",3600000)}globalThis.AI_SDK_LOG_WARNINGS??=!1;function Io(o,n){Wo(n)}async function Wo(o){if(E()){_.debug("[OpenAI] Skipping API validation in browser environment");return}if(!J(o)){_.warn("[OpenAI] OPENAI_API_KEY is not configured. OpenAI functionality will fail until a valid API key is provided.");return}try{let c=O(o),r=await fetch(`${c}/models`,{headers:R(o)});if(!r.ok){_.warn(`[OpenAI] API key validation failed: ${r.status} ${r.statusText}. Please verify your OPENAI_API_KEY is correct.`);return}}catch(c){let r=c instanceof Error?c.message:String(c);_.warn(`[OpenAI] API validation error: ${r}. OpenAI functionality may be limited.`)}}import{logger as U}from"@elizaos/core";import{logger as Xo}from"@elizaos/core";var S={WAV:{HEADER:[82,73,70,70],IDENTIFIER:[87,65,86,69]},MP3_ID3:[73,68,51],OGG:[79,103,103,83],FLAC:[102,76,97,67],FTYP:[102,116,121,112],WEBM_EBML:[26,69,223,163]},Co=12;function j(o,n,c){for(let r=0;r<c.length;r++){let f=c[r];if(f===void 0||o[n+r]!==f)return!1}return!0}function W(o){if(o.length<Co)return"application/octet-stream";if(j(o,0,S.WAV.HEADER)&&j(o,8,S.WAV.IDENTIFIER))return"audio/wav";let n=o[0],c=o[1];if(j(o,0,S.MP3_ID3)||n===255&&c!==void 0&&(c&224)===224)return"audio/mpeg";if(j(o,0,S.OGG))return"audio/ogg";if(j(o,0,S.FLAC))return"audio/flac";if(j(o,4,S.FTYP))return"audio/mp4";if(j(o,0,S.WEBM_EBML))return"audio/webm";return Xo.warn("Could not detect audio format from buffer, using generic binary type"),"application/octet-stream"}function Ho(o){switch(o){case"audio/wav":return"wav";case"audio/mpeg":return"mp3";case"audio/ogg":return"ogg";case"audio/flac":return"flac";case"audio/mp4":return"m4a";case"audio/webm":return"webm";case"application/octet-stream":return"bin"}}function po(o){return`recording.${Ho(o)}`}function wo(o){return o instanceof Blob||o instanceof File}function X(o){return Buffer.isBuffer(o)}function Vo(o){return typeof o==="object"&&o!==null&&"audio"in o&&(wo(o.audio)||X(o.audio))}function Qo(o){return typeof o==="object"&&o!==null&&"audioUrl"in o&&typeof o.audioUrl==="string"}async function so(o){let n=await fetch(o);if(!n.ok)throw Error(`Failed to fetch audio from URL: ${n.status}`);return n.blob()}async function C(o,n){let c=no(o),r,f={};if(typeof n==="string")U.debug(`[OpenAI] Fetching audio from URL: ${n}`),r=await so(n);else if(wo(n))r=n;else if(X(n)){let P=W(n);U.debug(`[OpenAI] Auto-detected audio MIME type: ${P}`),r=new Blob([new Uint8Array(n)],{type:P})}else if(Vo(n)){if(f=n,n.model)c=n.model;if(X(n.audio)){let P=n.mimeType??W(n.audio);U.debug(`[OpenAI] Using MIME type: ${P}`),r=new Blob([new Uint8Array(n.audio)],{type:P})}else r=n.audio}else if(Qo(n))U.debug(`[OpenAI] Fetching audio from URL: ${n.audioUrl}`),r=await so(n.audioUrl),f={prompt:n.prompt};else throw Error("TRANSCRIPTION expects Blob, File, Buffer, URL string, or TranscriptionParams object");U.debug(`[OpenAI] Using TRANSCRIPTION model: ${c}`);let T=r.type||"audio/webm",A=r.name||po(T.startsWith("audio/")?T:"audio/webm"),k=new FormData;if(k.append("file",r,A),k.append("model",c),f.language)k.append("language",f.language);if(f.responseFormat)k.append("response_format",f.responseFormat);if(f.prompt)k.append("prompt",f.prompt);if(f.temperature!==void 0)k.append("temperature",String(f.temperature));if(f.timestampGranularities)for(let P of f.timestampGranularities)k.append("timestamp_granularities[]",P);let t=O(o),i=await fetch(`${t}/audio/transcriptions`,{method:"POST",headers:R(o),body:k});if(!i.ok){let P=await i.text().catch(()=>"Unknown error");throw Error(`OpenAI transcription failed: ${i.status} ${i.statusText} - ${P}`)}return(await i.json()).text}async function H(o,n){let c,r,f="mp3",T,A;if(typeof n==="string")c=n,r=void 0;else{if(c=n.text,r=n.voice,"format"in n&&n.format)f=n.format;if("model"in n&&n.model)T=n.model;if("instructions"in n&&n.instructions)A=n.instructions}if(T=T??co(o),r=r??ro(o),A=A??fo(o),U.debug(`[OpenAI] Using TEXT_TO_SPEECH model: ${T}`),!c||c.trim().length===0)throw Error("TEXT_TO_SPEECH requires non-empty text");if(c.length>4096)throw Error("TEXT_TO_SPEECH text exceeds 4096 character limit");let k=["alloy","echo","fable","onyx","nova","shimmer"];if(r&&!k.includes(r))throw Error(`Invalid voice: ${r}. Must be one of: ${k.join(", ")}`);let t=O(o),i={model:T,voice:r,input:c,response_format:f};if(A&&A.length>0)i.instructions=A;let p=await fetch(`${t}/audio/speech`,{method:"POST",headers:{...R(o),"Content-Type":"application/json",...f==="mp3"?{Accept:"audio/mpeg"}:{}},body:JSON.stringify(i)});if(!p.ok){let P=await p.text().catch(()=>"Unknown error");throw Error(`OpenAI TTS failed: ${p.status} ${p.statusText} - ${P}`)}return p.arrayBuffer()}import{logger as x,ModelType as Go,VECTOR_DIMS as qo}from"@elizaos/core";import{EventType as Do}from"@elizaos/core";var Oo=200;function Zo(o){if(o.length<=Oo)return o;return`${o.slice(0,Oo)}…`}function Yo(o){if("promptTokens"in o){let n="promptTokensDetails"in o?o.promptTokensDetails:void 0,c=o.cachedPromptTokens??n?.cachedTokens;return{promptTokens:o.promptTokens??0,completionTokens:o.completionTokens??0,totalTokens:o.totalTokens??(o.promptTokens??0)+(o.completionTokens??0),cachedPromptTokens:c}}if("inputTokens"in o||"outputTokens"in o){let n=o.inputTokens??0,c=o.outputTokens??0,r=o.totalTokens??n+c;return{promptTokens:n,completionTokens:c,totalTokens:r,cachedPromptTokens:o.cachedInputTokens}}return{promptTokens:0,completionTokens:0,totalTokens:0}}function $(o,n,c,r){let f=Yo(r),T={runtime:o,source:"openai",provider:"openai",type:n,prompt:Zo(c),tokens:{prompt:f.promptTokens,completion:f.completionTokens,total:f.totalTokens,...f.cachedPromptTokens!==void 0?{cached:f.cachedPromptTokens}:{}}};o.emitEvent(Do.MODEL_USED,T)}var ho=8000;function Mo(o){let n=Object.values(qo);if(!n.includes(o))throw Error(`Invalid embedding dimension: ${o}. Must be one of: ${n.join(", ")}`);return o}function eo(o){if(o===null)return null;if(typeof o==="string")return o;if(typeof o==="object"&&typeof o.text==="string")return o.text;throw Error("Invalid embedding params: expected string, { text: string }, or null")}async function V(o,n){let c=m(o),r=Mo(ko(o)),f=eo(n);if(f===null){x.debug("[OpenAI] Creating test embedding for initialization");let y=Array(r).fill(0);return y[0]=0.1,y}let T=f.trim();if(T.length===0)throw Error("Cannot generate embedding for empty text");let A=ho*4;if(T.length>A)x.warn(`[OpenAI] Embedding input too long (~${Math.ceil(T.length/4)} tokens), truncating to ~${ho} tokens`),T=T.slice(0,A);let t=`${u(o)}/embeddings`;x.debug(`[OpenAI] Generating embedding with model: ${c}`);let i=await fetch(t,{method:"POST",headers:{...R(o,!0),"Content-Type":"application/json"},body:JSON.stringify({model:c,input:T})});if(!i.ok){let y=await i.text().catch(()=>"Unknown error");throw Error(`OpenAI embedding API error: ${i.status} ${i.statusText} - ${y}`)}let p=await i.json(),P=p?.data?.[0];if(!P||!P.embedding)throw Error("OpenAI API returned invalid embedding response structure");let w=P.embedding;if(w.length!==r)throw Error(`Embedding dimension mismatch: got ${w.length}, expected ${r}. Check OPENAI_EMBEDDING_DIMENSIONS setting.`);if(p.usage)$(o,Go.TEXT_EMBEDDING,T,{promptTokens:p.usage.prompt_tokens,completionTokens:0,totalTokens:p.usage.total_tokens});return x.debug(`[OpenAI] Generated embedding with ${w.length} dimensions`),w}import{logger as lo,ModelType as Bo}from"@elizaos/core";var Ro="Please analyze this image and provide a title and detailed description.";async function Q(o,n){let c=To(o),r=n.count??1,f=n.size??"1024x1024",T=n;if(lo.debug(`[OpenAI] Using IMAGE model: ${c}`),!n.prompt||n.prompt.trim().length===0)throw Error("IMAGE generation requires a non-empty prompt");if(r<1||r>10)throw Error("IMAGE count must be between 1 and 10");let A=O(o),k={model:c,prompt:n.prompt,n:r,size:f};if(T.quality)k.quality=T.quality;if(T.style)k.style=T.style;let t=await fetch(`${A}/images/generations`,{method:"POST",headers:{...R(o),"Content-Type":"application/json"},body:JSON.stringify(k)});if(!t.ok){let p=await t.text().catch(()=>"Unknown error");throw Error(`OpenAI image generation failed: ${t.status} ${t.statusText} - ${p}`)}let i=await t.json();if(!i.data||i.data.length===0)throw Error("OpenAI API returned no images");return i.data.map((p)=>({url:p.url,revisedPrompt:p.revised_prompt}))}function ao(o){return o.match(/title[:\s]+(.+?)(?:\n|$)/i)?.[1]?.trim()??"Image Analysis"}function vo(o){return o.replace(/title[:\s]+(.+?)(?:\n|$)/i,"").trim()}async function D(o,n){let c=oo(o),r=Ao(o);lo.debug(`[OpenAI] Using IMAGE_DESCRIPTION model: ${c}`);let f,T;if(typeof n==="string")f=n,T=Ro;else f=n.imageUrl,T=n.prompt??Ro;if(!f||f.trim().length===0)throw Error("IMAGE_DESCRIPTION requires a valid image URL");let A=O(o),k={model:c,messages:[{role:"user",content:[{type:"text",text:T},{type:"image_url",image_url:{url:f}}]}],max_tokens:r},t=await fetch(`${A}/chat/completions`,{method:"POST",headers:{...R(o),"Content-Type":"application/json"},body:JSON.stringify(k)});if(!t.ok){let w=await t.text().catch(()=>"Unknown error");throw Error(`OpenAI image description failed: ${t.status} ${t.statusText} - ${w}`)}let i=await t.json();if(i.usage)$(o,Bo.IMAGE_DESCRIPTION,typeof n==="string"?n:n.prompt??"",{promptTokens:i.usage.prompt_tokens,completionTokens:i.usage.completion_tokens,totalTokens:i.usage.total_tokens});let P=i.choices?.[0]?.message?.content;if(!P)throw Error("OpenAI API returned empty image description");return{title:ao(P),description:vo(P)}}import{logger as bo,ModelType as No}from"@elizaos/core";import{generateObject as on}from"ai";import{createOpenAI as yo}from"@ai-sdk/openai";var go="sk-proxy";function L(o){let n=O(o),c=J(o);if(!c&&g(o))return yo({apiKey:go,baseURL:n});if(!c)throw Error("OPENAI_API_KEY is required. Set it in your environment variables or runtime settings.");return yo({apiKey:c,baseURL:n})}import{logger as Jo}from"@elizaos/core";import{JSONParseError as uo}from"ai";var mo={MARKDOWN_JSON:/```json\n|\n```|```/g,WHITESPACE:/^\s+|\s+$/g};function $o(){return async({text:o,error:n})=>{if(!(n instanceof uo))return null;try{let c=o.replace(mo.MARKDOWN_JSON,"");return JSON.parse(c),Jo.debug("[JSON Repair] Successfully repaired JSON by removing markdown wrappers"),c}catch{return Jo.warn("[JSON Repair] Unable to repair JSON text"),null}}}async function So(o,n,c,r){let f=L(o),T=r(o);if(bo.debug(`[OpenAI] Using ${c} model: ${T}`),!n.prompt||n.prompt.trim().length===0)throw Error("Object generation requires a non-empty prompt");if(n.schema)bo.debug("[OpenAI] Schema provided but using no-schema mode. Structure is determined by prompt instructions.");let A=f.chat(T),{object:k,usage:t}=await on({model:A,output:"no-schema",prompt:n.prompt,experimental_repairText:$o()});if(t)$(o,c,n.prompt,t);if(typeof k!=="object"||k===null)throw Error(`Object generation returned ${typeof k}, expected object`);return k}async function Z(o,n){return So(o,n,No.OBJECT_SMALL,F)}async function Y(o,n){return So(o,n,No.OBJECT_LARGE,d)}import{logger as b}from"@elizaos/core";function nn(o){switch(o.type){case"web_search_preview":return{type:"web_search_preview"};case"file_search":return{type:"file_search",vector_store_ids:o.vectorStoreIds};case"code_interpreter":return{type:"code_interpreter",container:o.container??{type:"auto"}};case"mcp":return{type:"mcp",server_label:o.serverLabel,server_url:o.serverUrl,require_approval:o.requireApproval??"never"};default:throw Error(`Unknown research tool type: ${o.type}`)}}function cn(o){switch(o.type){case"web_search_call":return{id:o.id??"",type:"web_search_call",status:o.status??"completed",action:{type:o.action?.type??"search",query:o.action?.query,url:o.action?.url}};case"file_search_call":return{id:o.id??"",type:"file_search_call",status:o.status??"completed",query:o.query??"",results:o.results?.map((n)=>({fileId:n.file_id,fileName:n.file_name,score:n.score}))};case"code_interpreter_call":return{id:o.id??"",type:"code_interpreter_call",status:o.status??"completed",code:o.code??"",output:o.output};case"mcp_tool_call":return{id:o.id??"",type:"mcp_tool_call",status:o.status??"completed",serverLabel:o.server_label??"",toolName:o.tool_name??"",arguments:o.arguments??{},result:o.result};case"message":return{type:"message",content:o.content?.map((n)=>({type:"output_text",text:n.text,annotations:n.annotations?.map((c)=>({url:c.url,title:c.title,startIndex:c.start_index,endIndex:c.end_index}))??[]}))??[]};default:return null}}function rn(o){if(o.output_text){let r=[];if(o.output){for(let f of o.output)if(f.type==="message"&&f.content){for(let T of f.content)if(T.annotations)for(let A of T.annotations)r.push({url:A.url,title:A.title,startIndex:A.start_index,endIndex:A.end_index})}}return{text:o.output_text,annotations:r}}let n="",c=[];if(o.output){for(let r of o.output)if(r.type==="message"&&r.content){for(let f of r.content)if(n+=f.text,f.annotations)for(let T of f.annotations)c.push({url:T.url,title:T.title,startIndex:T.start_index,endIndex:T.end_index})}}return{text:n,annotations:c}}async function G(o,n){let c=J(o);if(!c)throw Error("OPENAI_API_KEY is required for deep research. Set it in your environment variables or runtime settings.");let r=O(o),f=n.model??to(o),T=Po(o);b.debug(`[OpenAI] Starting deep research with model: ${f}`),b.debug(`[OpenAI] Research input: ${n.input.substring(0,100)}...`);let A=n.tools?.filter((N)=>N.type==="web_search_preview"||N.type==="file_search"||N.type==="mcp");if(!A||A.length===0)b.debug("[OpenAI] No data source tools specified, defaulting to web_search_preview"),n.tools=[{type:"web_search_preview"},...n.tools??[]];let k={model:f,input:n.input};if(n.instructions)k.instructions=n.instructions;if(n.background!==void 0)k.background=n.background;if(n.tools&&n.tools.length>0)k.tools=n.tools.map(nn);if(n.maxToolCalls!==void 0)k.max_tool_calls=n.maxToolCalls;if(n.reasoningSummary)k.reasoning={summary:n.reasoningSummary};b.debug(`[OpenAI] Research request body: ${JSON.stringify(k,null,2)}`);let t=await fetch(`${r}/responses`,{method:"POST",headers:{Authorization:`Bearer ${c}`,"Content-Type":"application/json"},body:JSON.stringify(k),signal:AbortSignal.timeout(T)});if(!t.ok){let N=await t.text();throw b.error(`[OpenAI] Research request failed: ${t.status} ${N}`),Error(`Deep research request failed: ${t.status} ${t.statusText}`)}let i=await t.json();if(i.error)throw b.error(`[OpenAI] Research API error: ${i.error.message}`),Error(`Deep research error: ${i.error.message}`);b.debug(`[OpenAI] Research response received. Status: ${i.status??"completed"}`);let{text:p,annotations:P}=rn(i),w=[];if(i.output)for(let N of i.output){let a=cn(N);if(a)w.push(a)}let y={id:i.id,text:p,annotations:P,outputItems:w,status:i.status};return b.info(`[OpenAI] Research completed. Text length: ${p.length}, Annotations: ${P.length}, Output items: ${w.length}`),y}import{logger as fn,ModelType as jo}from"@elizaos/core";import{generateText as Tn,streamText as kn}from"ai";function An(o){if(!o)return;let n=o.inputTokens??0,c=o.outputTokens??0,r=o;return{promptTokens:n,completionTokens:c,totalTokens:n+c,cachedPromptTokens:r.cachedInputTokens}}function tn(o){let n=o;return{promptCacheKey:n.providerOptions?.openai?.promptCacheKey,promptCacheRetention:n.providerOptions?.openai?.promptCacheRetention}}async function Eo(o,n,c,r){let f=L(o),T=r(o);fn.debug(`[OpenAI] Using ${c} model: ${T}`);let A=tn(n),k=o.character.system??void 0,i={model:f.chat(T),prompt:n.prompt,system:k,maxOutputTokens:n.maxTokens??8192,experimental_telemetry:{isEnabled:io(o)},...A.promptCacheKey||A.promptCacheRetention?{providerOptions:{openai:{...A.promptCacheKey?{promptCacheKey:A.promptCacheKey}:{},...A.promptCacheRetention?{promptCacheRetention:A.promptCacheRetention}:{}}}}:{}};if(n.stream){let w=kn(i);return{textStream:w.textStream,text:Promise.resolve(w.text),usage:Promise.resolve(w.usage).then(An),finishReason:Promise.resolve(w.finishReason).then((y)=>y)}}let{text:p,usage:P}=await Tn(i);if(P)$(o,c,n.prompt,P);return p}async function q(o,n){return Eo(o,n,jo.TEXT_SMALL,F)}async function M(o,n){return Eo(o,n,jo.TEXT_LARGE,d)}import{ModelType as _o}from"@elizaos/core";import{ModelType as Pn}from"@elizaos/core";import{encodingForModel as In,getEncoding as pn}from"js-tiktoken";function Fo(o){let c=o.toLowerCase().includes("4o")?"o200k_base":"cl100k_base";try{return In(o)}catch{return pn(c)}}function Uo(o,n){if(n===Pn.TEXT_SMALL)return F(o);return d(o)}function zo(o,n,c){let r=Uo(o,n);return Fo(r).encode(c)}function Lo(o,n,c){let r=Uo(o,n);return Fo(r).decode(c)}async function e(o,n){if(!n.prompt)throw Error("Tokenization requires a non-empty prompt");let c=n.modelType??_o.TEXT_LARGE;return zo(o,c,n.prompt)}async function B(o,n){if(!n.tokens||!Array.isArray(n.tokens))throw Error("Detokenization requires a valid tokens array");if(n.tokens.length===0)return"";for(let r=0;r<n.tokens.length;r++){let f=n.tokens[r];if(typeof f!=="number"||!Number.isFinite(f))throw Error(`Invalid token at index ${r}: expected number`)}let c=n.modelType??_o.TEXT_LARGE;return Lo(o,c,n.tokens)}function sn(){if(typeof process>"u")return{};return process.env}var h=sn(),wn={name:"openai",description:"OpenAI API integration for text, image, audio, and embedding models",config:{OPENAI_API_KEY:h.OPENAI_API_KEY??null,OPENAI_BASE_URL:h.OPENAI_BASE_URL??null,OPENAI_SMALL_MODEL:h.OPENAI_SMALL_MODEL??null,OPENAI_LARGE_MODEL:h.OPENAI_LARGE_MODEL??null,SMALL_MODEL:h.SMALL_MODEL??null,LARGE_MODEL:h.LARGE_MODEL??null,OPENAI_EMBEDDING_MODEL:h.OPENAI_EMBEDDING_MODEL??null,OPENAI_EMBEDDING_API_KEY:h.OPENAI_EMBEDDING_API_KEY??null,OPENAI_EMBEDDING_URL:h.OPENAI_EMBEDDING_URL??null,OPENAI_EMBEDDING_DIMENSIONS:h.OPENAI_EMBEDDING_DIMENSIONS??null,OPENAI_IMAGE_DESCRIPTION_MODEL:h.OPENAI_IMAGE_DESCRIPTION_MODEL??null,OPENAI_IMAGE_DESCRIPTION_MAX_TOKENS:h.OPENAI_IMAGE_DESCRIPTION_MAX_TOKENS??null,OPENAI_EXPERIMENTAL_TELEMETRY:h.OPENAI_EXPERIMENTAL_TELEMETRY??null,OPENAI_RESEARCH_MODEL:h.OPENAI_RESEARCH_MODEL??null,OPENAI_RESEARCH_TIMEOUT:h.OPENAI_RESEARCH_TIMEOUT??null},async init(o,n){Io(o,n)},models:{[I.TEXT_EMBEDDING]:async(o,n)=>{return V(o,n)},[I.TEXT_TOKENIZER_ENCODE]:async(o,n)=>{return e(o,n)},[I.TEXT_TOKENIZER_DECODE]:async(o,n)=>{return B(o,n)},[I.TEXT_SMALL]:async(o,n)=>{return q(o,n)},[I.TEXT_LARGE]:async(o,n)=>{return M(o,n)},[I.IMAGE]:async(o,n)=>{return Q(o,n)},[I.IMAGE_DESCRIPTION]:async(o,n)=>{return D(o,n)},[I.TRANSCRIPTION]:async(o,n)=>{return C(o,n)},[I.TEXT_TO_SPEECH]:async(o,n)=>{return H(o,n)},[I.OBJECT_SMALL]:async(o,n)=>{return Z(o,n)},[I.OBJECT_LARGE]:async(o,n)=>{return Y(o,n)},[I.RESEARCH]:async(o,n)=>{return G(o,n)}},tests:[{name:"openai_plugin_tests",tests:[{name:"openai_test_api_connectivity",fn:async(o)=>{let n=O(o),c=await fetch(`${n}/models`,{headers:R(o)});if(!c.ok)throw Error(`API connectivity test failed: ${c.status} ${c.statusText}`);let r=await c.json();l.info(`[OpenAI Test] API connected. ${r.data?.length??0} models available.`)}},{name:"openai_test_text_embedding",fn:async(o)=>{let n=await o.useModel(I.TEXT_EMBEDDING,{text:"Hello, world!"});if(!Array.isArray(n)||n.length===0)throw Error("Embedding should return a non-empty array");l.info(`[OpenAI Test] Generated embedding with ${n.length} dimensions`)}},{name:"openai_test_text_small",fn:async(o)=>{let n=await o.useModel(I.TEXT_SMALL,{prompt:"Say hello in exactly 5 words."});if(typeof n!=="string"||n.length===0)throw Error("TEXT_SMALL should return non-empty string");l.info(`[OpenAI Test] TEXT_SMALL generated: "${n.substring(0,50)}..."`)}},{name:"openai_test_text_large",fn:async(o)=>{let n=await o.useModel(I.TEXT_LARGE,{prompt:"Explain quantum computing in 2 sentences."});if(typeof n!=="string"||n.length===0)throw Error("TEXT_LARGE should return non-empty string");l.info(`[OpenAI Test] TEXT_LARGE generated: "${n.substring(0,50)}..."`)}},{name:"openai_test_tokenizer_roundtrip",fn:async(o)=>{let c=await o.useModel(I.TEXT_TOKENIZER_ENCODE,{prompt:"Hello, tokenizer test!",modelType:I.TEXT_SMALL});if(!Array.isArray(c)||c.length===0)throw Error("Tokenization should return non-empty token array");let r=await o.useModel(I.TEXT_TOKENIZER_DECODE,{tokens:c,modelType:I.TEXT_SMALL});if(r!=="Hello, tokenizer test!")throw Error(`Tokenizer roundtrip failed: expected "Hello, tokenizer test!", got "${r}"`);l.info(`[OpenAI Test] Tokenizer roundtrip successful (${c.length} tokens)`)}},{name:"openai_test_streaming",fn:async(o)=>{let n=[],c=await o.useModel(I.TEXT_LARGE,{prompt:"Count from 1 to 5, one number per line.",stream:!0,onStreamChunk:(r)=>{n.push(r)}});if(typeof c!=="string"||c.length===0)throw Error("Streaming should return non-empty result");if(n.length===0)throw Error("No streaming chunks received");l.info(`[OpenAI Test] Streaming test: ${n.length} chunks received`)}},{name:"openai_test_image_description",fn:async(o)=>{let c=await o.useModel(I.IMAGE_DESCRIPTION,"https://upload.wikimedia.org/wikipedia/commons/thumb/a/a7/Camponotus_flavomarginatus_ant.jpg/440px-Camponotus_flavomarginatus_ant.jpg");if(!c||typeof c!=="object"||!("title"in c)||!("description"in c))throw Error("Image description should return { title, description }");l.info(`[OpenAI Test] Image described: "${c.title}"`)}},{name:"openai_test_transcription",fn:async(o)=>{let r=await(await fetch("https://upload.wikimedia.org/wikipedia/commons/2/25/En-Open_Source.ogg")).arrayBuffer(),f=Buffer.from(new Uint8Array(r)),T=await o.useModel(I.TRANSCRIPTION,f);if(typeof T!=="string")throw Error("Transcription should return a string");l.info(`[OpenAI Test] Transcription: "${T.substring(0,50)}..."`)}},{name:"openai_test_text_to_speech",fn:async(o)=>{let n=await o.useModel(I.TEXT_TO_SPEECH,{text:"Hello, this is a text-to-speech test."});if(!(n instanceof ArrayBuffer)||n.byteLength===0)throw Error("TTS should return non-empty ArrayBuffer");l.info(`[OpenAI Test] TTS generated ${n.byteLength} bytes of audio`)}},{name:"openai_test_object_generation",fn:async(o)=>{let n=await o.useModel(I.OBJECT_SMALL,{prompt:"Return a JSON object with exactly these fields: name (string), age (number), active (boolean)"});if(!n||typeof n!=="object")throw Error("Object generation should return an object");l.info(`[OpenAI Test] Object generated: ${JSON.stringify(n).substring(0,100)}`)}},{name:"openai_test_research",fn:async(o)=>{let n=await o.useModel(I.RESEARCH,{input:"What is the current date and time?",tools:[{type:"web_search_preview"}],maxToolCalls:3});if(!n||typeof n!=="object"||!("text"in n))throw Error("Research should return an object with text property");if(typeof n.text!=="string"||n.text.length===0)throw Error("Research result text should be a non-empty string");l.info(`[OpenAI Test] Research completed. Text length: ${n.text.length}, Annotations: ${n.annotations?.length??0}`)}}]}]},On=wn;export{wn as openaiPlugin,On as default};
-
-//# debugId=01A431D38E4333C464756E2164756E21