cowork-os 0.3.21 → 0.3.25

package/dist/electron/shared/llm-provider-catalog.js

@@ -0,0 +1,198 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.CUSTOM_PROVIDER_IDS = exports.CUSTOM_PROVIDER_MAP = exports.CUSTOM_PROVIDER_CATALOG = void 0;
+exports.CUSTOM_PROVIDER_CATALOG = [
+    {
+        id: 'opencode',
+        name: 'OpenCode Zen',
+        compatibility: 'openai',
+        defaultModel: '',
+        apiKeyLabel: 'API Key / Token',
+        apiKeyPlaceholder: 'Enter token',
+        requiresBaseUrl: true,
+        description: 'OpenAI-compatible endpoint required.',
+    },
+    {
+        id: 'google-vertex',
+        name: 'Google Vertex',
+        compatibility: 'openai',
+        defaultModel: 'gemini-2.0-flash',
+        apiKeyLabel: 'Access Token',
+        apiKeyPlaceholder: 'ya29...',
+        requiresBaseUrl: true,
+        description: 'Requires a compatible gateway endpoint.',
+    },
+    {
+        id: 'google-antigravity',
+        name: 'Google Antigravity',
+        compatibility: 'openai',
+        defaultModel: 'gemini-2.0-flash',
+        apiKeyLabel: 'Access Token',
+        apiKeyPlaceholder: 'ya29...',
+        requiresBaseUrl: true,
+        description: 'Requires a compatible gateway endpoint.',
+    },
+    {
+        id: 'google-gemini-cli',
+        name: 'Google Gemini CLI',
+        compatibility: 'openai',
+        defaultModel: 'gemini-2.0-flash',
+        apiKeyLabel: 'Access Token',
+        apiKeyPlaceholder: 'ya29...',
+        requiresBaseUrl: true,
+        description: 'Requires a compatible gateway endpoint.',
+    },
+    {
+        id: 'zai',
+        name: 'Z.AI',
+        compatibility: 'openai',
+        defaultModel: 'glm-4.7',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'zai-...',
+        requiresBaseUrl: true,
+        description: 'OpenAI-compatible endpoint required.',
+    },
+    {
+        id: 'glm',
+        name: 'GLM',
+        compatibility: 'openai',
+        defaultModel: 'glm-4-plus',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'glm-...',
+        requiresBaseUrl: true,
+        description: 'OpenAI-compatible endpoint required.',
+    },
+    {
+        id: 'vercel-ai-gateway',
+        name: 'Vercel AI Gateway',
+        compatibility: 'anthropic',
+        baseUrl: 'https://ai-gateway.vercel.sh/v1',
+        defaultModel: 'claude-3-5-sonnet-20241022',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'vgw_...',
+        apiKeyUrl: 'https://vercel.com/docs/ai-gateway',
+    },
+    {
+        id: 'moonshot',
+        name: 'Moonshot (Kimi)',
+        compatibility: 'openai',
+        baseUrl: 'https://api.moonshot.ai/v1',
+        defaultModel: 'kimi-k2.5',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'sk-...',
+    },
+    {
+        id: 'cerebras',
+        name: 'Cerebras',
+        compatibility: 'openai',
+        baseUrl: 'https://api.cerebras.ai/v1',
+        defaultModel: 'llama3.1-8b',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'csk_...',
+        apiKeyUrl: 'https://cloud.cerebras.ai/',
+    },
+    {
+        id: 'mistral',
+        name: 'Mistral',
+        compatibility: 'openai',
+        baseUrl: 'https://api.mistral.ai/v1',
+        defaultModel: 'mistral-large-latest',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'mistral-...',
+        apiKeyUrl: 'https://console.mistral.ai/',
+    },
+    {
+        id: 'github-copilot',
+        name: 'GitHub Copilot',
+        compatibility: 'openai',
+        defaultModel: 'gpt-4o',
+        apiKeyLabel: 'GitHub Token',
+        apiKeyPlaceholder: 'ghp_...',
+    },
+    {
+        id: 'qwen-portal',
+        name: 'Qwen',
+        compatibility: 'anthropic',
+        baseUrl: 'https://portal.qwen.ai/v1',
+        defaultModel: 'coder-model',
+        apiKeyLabel: 'API Key / Token',
+        apiKeyPlaceholder: 'qwen-...',
+    },
+    {
+        id: 'minimax',
+        name: 'MiniMax',
+        compatibility: 'openai',
+        baseUrl: 'https://api.minimax.chat/v1',
+        defaultModel: 'MiniMax-M2.1',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'minimax-...',
+    },
+    {
+        id: 'minimax-portal',
+        name: 'MiniMax Portal',
+        compatibility: 'anthropic',
+        baseUrl: 'https://api.minimax.io/anthropic',
+        defaultModel: 'MiniMax-M2.1',
+        apiKeyLabel: 'API Key / Token',
+        apiKeyPlaceholder: 'minimax-...',
+    },
+    {
+        id: 'xiaomi',
+        name: 'Xiaomi MiMo',
+        compatibility: 'anthropic',
+        baseUrl: 'https://api.xiaomimimo.com/anthropic',
+        defaultModel: 'mimo-v2-flash',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'mimo-...',
+    },
+    {
+        id: 'venice',
+        name: 'Venice AI',
+        compatibility: 'openai',
+        baseUrl: 'https://api.venice.ai/api/v1',
+        defaultModel: 'llama-3.3-70b',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'venice-...',
+    },
+    {
+        id: 'synthetic',
+        name: 'Synthetic',
+        compatibility: 'anthropic',
+        baseUrl: 'https://api.synthetic.new/anthropic',
+        defaultModel: 'hf:MiniMaxAI/MiniMax-M2.1',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'syn_...',
+    },
+    {
+        id: 'kimi-code',
+        name: 'Kimi Code',
+        compatibility: 'openai',
+        baseUrl: 'https://api.kimi.com/coding/v1',
+        defaultModel: 'kimi-for-coding',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'sk-...',
+    },
+    {
+        id: 'openai-compatible',
+        name: 'OpenAI-Compatible (Custom)',
+        compatibility: 'openai',
+        baseUrl: 'http://localhost:1234/v1',
+        defaultModel: 'gpt-4o-mini',
+        apiKeyLabel: 'API Key (Optional)',
+        apiKeyPlaceholder: 'sk-...',
+        requiresBaseUrl: true,
+        apiKeyOptional: true,
+    },
+    {
+        id: 'anthropic-compatible',
+        name: 'Anthropic-Compatible (Custom)',
+        compatibility: 'anthropic',
+        baseUrl: 'http://localhost:4000',
+        defaultModel: 'claude-3-5-sonnet-20241022',
+        apiKeyLabel: 'API Key',
+        apiKeyPlaceholder: 'sk-...',
+        requiresBaseUrl: true,
+    },
+];
+exports.CUSTOM_PROVIDER_MAP = new Map(exports.CUSTOM_PROVIDER_CATALOG.map((provider) => [provider.id, provider]));
+exports.CUSTOM_PROVIDER_IDS = new Set(exports.CUSTOM_PROVIDER_CATALOG.map((provider) => provider.id));

package/dist/electron/shared/types.js

@@ -1,7 +1,7 @@
 "use strict";
 // Core types shared between main and renderer processes
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.VOICE_LANGUAGES = exports.OPENAI_VOICES = exports.DEFAULT_VOICE_SETTINGS = exports.ANALOGY_DOMAINS = exports.DEFAULT_RELATIONSHIP = exports.DEFAULT_QUIRKS = exports.DEFAULT_RESPONSE_STYLE = exports.PERSONA_DEFINITIONS = exports.PERSONALITY_DEFINITIONS = exports.DEFAULT_QUEUE_SETTINGS = exports.DEFAULT_BLOCKED_COMMAND_PATTERNS = exports.DEFAULT_TRUSTED_COMMAND_PATTERNS = exports.IPC_CHANNELS = exports.BOARD_COLUMNS = exports.DEFAULT_AGENT_ROLES = exports.TEMP_WORKSPACE_NAME = exports.TEMP_WORKSPACE_ID = exports.CONTEXT_TOOL_RESTRICTIONS = exports.TOOL_RISK_LEVELS = exports.TOOL_GROUPS = exports.ACCENT_COLORS = void 0;
+exports.VOICE_LANGUAGES = exports.OPENAI_VOICES = exports.DEFAULT_VOICE_SETTINGS = exports.ANALOGY_DOMAINS = exports.DEFAULT_RELATIONSHIP = exports.DEFAULT_QUIRKS = exports.DEFAULT_RESPONSE_STYLE = exports.PERSONA_DEFINITIONS = exports.PERSONALITY_DEFINITIONS = exports.DEFAULT_QUEUE_SETTINGS = exports.DEFAULT_BLOCKED_COMMAND_PATTERNS = exports.DEFAULT_TRUSTED_COMMAND_PATTERNS = exports.LLM_PROVIDER_TYPES = exports.CUSTOM_LLM_PROVIDER_TYPES = exports.BUILTIN_LLM_PROVIDER_TYPES = exports.IPC_CHANNELS = exports.BOARD_COLUMNS = exports.DEFAULT_AGENT_ROLES = exports.TEMP_WORKSPACE_NAME = exports.TEMP_WORKSPACE_ID = exports.CONTEXT_TOOL_RESTRICTIONS = exports.TOOL_RISK_LEVELS = exports.TOOL_GROUPS = exports.ACCENT_COLORS = void 0;
 exports.getPersonalityById = getPersonalityById;
 exports.getPersonaById = getPersonaById;
 exports.ACCENT_COLORS = [
@@ -59,6 +59,12 @@ exports.TOOL_GROUPS = {
     'group:network': [
         'web_search',
         'x_action',
+        'notion_action',
+        'box_action',
+        'onedrive_action',
+        'google_drive_action',
+        'dropbox_action',
+        'sharepoint_action',
         'browser_navigate',
         'browser_screenshot',
         'browser_get_content',
@@ -141,6 +147,12 @@ exports.TOOL_RISK_LEVELS = {
     browser_save_pdf: 'network',
     browser_close: 'network',
     x_action: 'network',
+    notion_action: 'network',
+    box_action: 'network',
+    onedrive_action: 'network',
+    google_drive_action: 'network',
+    dropbox_action: 'network',
+    sharepoint_action: 'network',
     // Meta
     revise_plan: 'read',
 };
@@ -462,6 +474,7 @@ exports.IPC_CHANNELS = {
     WORKSPACE_LIST: 'workspace:list',
     WORKSPACE_CREATE: 'workspace:create',
     WORKSPACE_UPDATE_PERMISSIONS: 'workspace:updatePermissions',
+    WORKSPACE_TOUCH: 'workspace:touch',
     WORKSPACE_GET_TEMP: 'workspace:getTemp', // Get or create temp workspace
     // Approval operations
     APPROVAL_RESPOND: 'approval:respond',
@@ -503,6 +516,9 @@ exports.IPC_CHANNELS = {
     LLM_GET_GEMINI_MODELS: 'llm:getGeminiModels',
     LLM_GET_OPENROUTER_MODELS: 'llm:getOpenRouterModels',
     LLM_GET_OPENAI_MODELS: 'llm:getOpenAIModels',
+    LLM_GET_GROQ_MODELS: 'llm:getGroqModels',
+    LLM_GET_XAI_MODELS: 'llm:getXAIModels',
+    LLM_GET_KIMI_MODELS: 'llm:getKimiModels',
     LLM_OPENAI_OAUTH_START: 'llm:openaiOAuthStart',
     LLM_OPENAI_OAUTH_LOGOUT: 'llm:openaiOAuthLogout',
     LLM_GET_BEDROCK_MODELS: 'llm:getBedrockModels',
@@ -528,6 +544,36 @@ exports.IPC_CHANNELS = {
     X_SAVE_SETTINGS: 'x:saveSettings',
     X_TEST_CONNECTION: 'x:testConnection',
     X_GET_STATUS: 'x:getStatus',
+    // Notion Settings
+    NOTION_GET_SETTINGS: 'notion:getSettings',
+    NOTION_SAVE_SETTINGS: 'notion:saveSettings',
+    NOTION_TEST_CONNECTION: 'notion:testConnection',
+    NOTION_GET_STATUS: 'notion:getStatus',
+    // Box Settings
+    BOX_GET_SETTINGS: 'box:getSettings',
+    BOX_SAVE_SETTINGS: 'box:saveSettings',
+    BOX_TEST_CONNECTION: 'box:testConnection',
+    BOX_GET_STATUS: 'box:getStatus',
+    // OneDrive Settings
+    ONEDRIVE_GET_SETTINGS: 'onedrive:getSettings',
+    ONEDRIVE_SAVE_SETTINGS: 'onedrive:saveSettings',
+    ONEDRIVE_TEST_CONNECTION: 'onedrive:testConnection',
+    ONEDRIVE_GET_STATUS: 'onedrive:getStatus',
+    // Google Drive Settings
+    GOOGLE_DRIVE_GET_SETTINGS: 'googleDrive:getSettings',
+    GOOGLE_DRIVE_SAVE_SETTINGS: 'googleDrive:saveSettings',
+    GOOGLE_DRIVE_TEST_CONNECTION: 'googleDrive:testConnection',
+    GOOGLE_DRIVE_GET_STATUS: 'googleDrive:getStatus',
+    // Dropbox Settings
+    DROPBOX_GET_SETTINGS: 'dropbox:getSettings',
+    DROPBOX_SAVE_SETTINGS: 'dropbox:saveSettings',
+    DROPBOX_TEST_CONNECTION: 'dropbox:testConnection',
+    DROPBOX_GET_STATUS: 'dropbox:getStatus',
+    // SharePoint Settings
+    SHAREPOINT_GET_SETTINGS: 'sharepoint:getSettings',
+    SHAREPOINT_SAVE_SETTINGS: 'sharepoint:saveSettings',
+    SHAREPOINT_TEST_CONNECTION: 'sharepoint:testConnection',
+    SHAREPOINT_GET_STATUS: 'sharepoint:getStatus',
     // App Updates
     APP_CHECK_UPDATES: 'app:checkUpdates',
     APP_DOWNLOAD_UPDATE: 'app:downloadUpdate',
@@ -579,6 +625,8 @@ exports.IPC_CHANNELS = {
     MCP_REGISTRY_UNINSTALL: 'mcp:registryUninstall',
     MCP_REGISTRY_CHECK_UPDATES: 'mcp:registryCheckUpdates',
     MCP_REGISTRY_UPDATE_SERVER: 'mcp:registryUpdateServer',
+    // MCP Connector OAuth
+    MCP_CONNECTOR_OAUTH_START: 'mcp:connectorOAuthStart',
     // MCP Host
     MCP_HOST_START: 'mcp:hostStart',
     MCP_HOST_STOP: 'mcp:hostStop',
@@ -671,6 +719,7 @@ exports.IPC_CHANNELS = {
     CANVAS_EXPORT_HTML: 'canvas:exportHTML',
     CANVAS_EXPORT_TO_FOLDER: 'canvas:exportToFolder',
     CANVAS_OPEN_IN_BROWSER: 'canvas:openInBrowser',
+    CANVAS_OPEN_URL: 'canvas:openUrl',
     CANVAS_GET_SESSION_DIR: 'canvas:getSessionDir',
     // Mobile Companion Nodes
     NODE_LIST: 'node:list',
@@ -718,6 +767,46 @@ exports.IPC_CHANNELS = {
     VOICE_TEST_AZURE: 'voice:testAzure',
     VOICE_EVENT: 'voice:event',
 };
+// LLM Provider types
+exports.BUILTIN_LLM_PROVIDER_TYPES = [
+    'anthropic',
+    'bedrock',
+    'ollama',
+    'gemini',
+    'openrouter',
+    'openai',
+    'azure',
+    'groq',
+    'xai',
+    'kimi',
+];
+exports.CUSTOM_LLM_PROVIDER_TYPES = [
+    'moonshot',
+    'opencode',
+    'google-vertex',
+    'google-antigravity',
+    'google-gemini-cli',
+    'zai',
+    'glm',
+    'vercel-ai-gateway',
+    'cerebras',
+    'mistral',
+    'github-copilot',
+    'qwen-portal',
+    'minimax',
+    'minimax-portal',
+    'xiaomi',
+    'venice',
+    'synthetic',
+    'kimi-code',
+    'kimi-coding',
+    'openai-compatible',
+    'anthropic-compatible',
+];
+exports.LLM_PROVIDER_TYPES = [
+    ...exports.BUILTIN_LLM_PROVIDER_TYPES,
+    ...exports.CUSTOM_LLM_PROVIDER_TYPES,
+];
 // Default trusted command patterns (glob-like patterns)
 exports.DEFAULT_TRUSTED_COMMAND_PATTERNS = [
     'npm test*',

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "cowork-os",
-  "version": "0.3.21",
+  "version": "0.3.25",
   "description": "CoWork OS - The operating system for personal AI assistants",
   "overrides": {
     "undici": "^7.0.0"
@@ -44,9 +44,10 @@
     "dev": "concurrently \"npm run dev:react\" \"npm run dev:electron\"",
     "dev:react": "vite",
     "dev:electron": "tsc -p tsconfig.electron.json && cross-env NODE_ENV=development electron .",
-    "build": "npm run build:react && npm run build:electron",
+    "build": "npm run build:react && npm run build:electron && npm run build:connectors",
     "build:react": "vite build",
     "build:electron": "tsc -p tsconfig.electron.json",
+    "build:connectors": "concurrently -n salesforce,jira,hubspot,zendesk,servicenow,linear,asana,okta \"tsc -p connectors/salesforce-mcp/tsconfig.json\" \"tsc -p connectors/jira-mcp/tsconfig.json\" \"tsc -p connectors/hubspot-mcp/tsconfig.json\" \"tsc -p connectors/zendesk-mcp/tsconfig.json\" \"tsc -p connectors/servicenow-mcp/tsconfig.json\" \"tsc -p connectors/linear-mcp/tsconfig.json\" \"tsc -p connectors/asana-mcp/tsconfig.json\" \"tsc -p connectors/okta-mcp/tsconfig.json\"",
     "package": "npm run build && electron-builder",
     "lint": "eslint src --ext ts,tsx",
     "type-check": "tsc --noEmit",
@@ -135,7 +136,17 @@
         "from": "resources/skills",
         "to": "skills",
         "filter": [
-          "**/*.json"
+          "**/*.json",
+          "scripts/**"
+        ]
+      },
+      {
+        "from": "connectors",
+        "to": "connectors",
+        "filter": [
+          "**/dist/**",
+          "**/README.md",
+          "**/package.json"
         ]
       }
     ],

package/resources/skills/nano-banana-pro.json

@@ -1,10 +1,10 @@
 {
   "id": "nano-banana-pro",
   "name": "Nano-banana-pro",
-  "description": "Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).",
-  "icon": "🍌",
+  "description": "Generate images using Nano Banana Pro (Gemini 3 Pro) via the built-in tool.",
+  "icon": "\ud83c\udf4c",
   "category": "Tools",
-  "prompt": "# Nano Banana Pro (Gemini 3 Pro Image)\n\nUse the bundled script to generate or edit images.\n\nGenerate\n\n```bash\nuv run {baseDir}/scripts/generate_image.py --prompt \"your image description\" --filename \"output.png\" --resolution 1K\n```\n\nEdit (single image)\n\n```bash\nuv run {baseDir}/scripts/generate_image.py --prompt \"edit instructions\" --filename \"output.png\" -i \"/path/in.png\" --resolution 2K\n```\n\nMulti-image composition (up to 14 images)\n\n```bash\nuv run {baseDir}/scripts/generate_image.py --prompt \"combine these into one scene\" --filename \"output.png\" -i img1.png -i img2.png -i img3.png\n```\n\nAPI key\n\n- `GEMINI_API_KEY` env var\n- Or set `skills.\"nano-banana-pro\".apiKey` / `skills.\"nano-banana-pro\".env.GEMINI_API_KEY` in `~/.CoWork-OSS/CoWork-OSS.json`\n\nNotes\n\n- Resolutions: `1K` (default), `2K`, `4K`.\n- Use timestamps in filenames: `yyyy-mm-dd-hh-mm-ss-name.png`.\n- The script prints a `MEDIA:` line for CoWork-OSS to auto-attach on supported chat providers.\n- Do not read the image back; report the saved path only.",
+  "prompt": "# Nano Banana Pro (Image Generation)\n\nUse the built-in image generation tool.\n\nGenerate (preferred)\n\n- Call the `generate_image` tool with:\n  - `model`: `nano-banana-pro`\n  - `prompt`: your image description\n  - `imageSize`: `1K` or `2K`\n  - `filename`: optional (saved into the workspace)\n\nAPI key\n\n- Requires Gemini API key configured in Settings.\n- If the key is missing, tell the user to set it up and do not attempt fallbacks.\n",
   "parameters": [],
   "enabled": true
-}
+}

package/resources/skills/openai-image-gen.json

@@ -2,9 +2,9 @@
   "id": "openai-image-gen",
   "name": "Openai-image-gen",
   "description": "Batch-generate images via OpenAI Images API. Random prompt sampler + `index.html` gallery.",
-  "icon": "🖼️",
+  "icon": "\ud83d\uddbc\ufe0f",
   "category": "Tools",
-  "prompt": "# OpenAI Image Gen\n\nGenerate a handful of “random but structured” prompts and render them via the OpenAI Images API.\n\n## Run\n\n```bash\npython3 {baseDir}/scripts/gen.py\nopen ~/Projects/tmp/openai-image-gen-*/index.html  # if ~/Projects/tmp exists; else ./tmp/...\n```\n\nUseful flags:\n\n```bash\n# GPT image models with various options\npython3 {baseDir}/scripts/gen.py --count 16 --model gpt-image-1\npython3 {baseDir}/scripts/gen.py --prompt \"ultra-detailed studio photo of a lobster astronaut\" --count 4\npython3 {baseDir}/scripts/gen.py --size 1536x1024 --quality high --out-dir ./out/images\npython3 {baseDir}/scripts/gen.py --model gpt-image-1.5 --background transparent --output-format webp\n\n# DALL-E 3 (note: count is automatically limited to 1)\npython3 {baseDir}/scripts/gen.py --model dall-e-3 --quality hd --size 1792x1024 --style vivid\npython3 {baseDir}/scripts/gen.py --model dall-e-3 --style natural --prompt \"serene mountain landscape\"\n\n# DALL-E 2\npython3 {baseDir}/scripts/gen.py --model dall-e-2 --size 512x512 --count 4\n```\n\n## Model-Specific Parameters\n\nDifferent models support different parameter values. The script automatically selects appropriate defaults based on the model.\n\n### Size\n\n- **GPT image models** (`gpt-image-1`, `gpt-image-1-mini`, `gpt-image-1.5`): `1024x1024`, `1536x1024` (landscape), `1024x1536` (portrait), or `auto`\n  - Default: `1024x1024`\n- **dall-e-3**: `1024x1024`, `1792x1024`, or `1024x1792`\n  - Default: `1024x1024`\n- **dall-e-2**: `256x256`, `512x512`, or `1024x1024`\n  - Default: `1024x1024`\n\n### Quality\n\n- **GPT image models**: `auto`, `high`, `medium`, or `low`\n  - Default: `high`\n- **dall-e-3**: `hd` or `standard`\n  - Default: `standard`\n- **dall-e-2**: `standard` only\n  - Default: `standard`\n\n### Other Notable Differences\n\n- **dall-e-3** only supports generating 1 image at a time (`n=1`). The script automatically limits count to 1 when using this model.\n- **GPT image models** support additional parameters:\n  - `--background`: `transparent`, `opaque`, or `auto` (default)\n  - `--output-format`: `png` (default), `jpeg`, or `webp`\n  - Note: `stream` and `moderation` are available via API but not yet implemented in this script\n- **dall-e-3** has a `--style` parameter: `vivid` (hyper-real, dramatic) or `natural` (more natural looking)\n\n## Output\n\n- `*.png`, `*.jpeg`, or `*.webp` images (output format depends on model + `--output-format`)\n- `prompts.json` (prompt → file mapping)\n- `index.html` (thumbnail gallery)",
+  "prompt": "# OpenAI Image Gen\n\nGenerate a handful of \u201crandom but structured\u201d prompts and render them via the OpenAI Images API.\n\n## Run\n\n```bash\npython3 {baseDir}/scripts/gen.py\nopen ~/Projects/tmp/openai-image-gen-*/index.html  # if ~/Projects/tmp exists; else ./tmp/...\n```\n\nUseful flags:\n\n```bash\n# GPT image models with various options\npython3 {baseDir}/scripts/gen.py --count 16 --model gpt-image-1\npython3 {baseDir}/scripts/gen.py --prompt \"ultra-detailed studio photo of a lobster astronaut\" --count 4\npython3 {baseDir}/scripts/gen.py --size 1536x1024 --quality high --out-dir ./out/images\npython3 {baseDir}/scripts/gen.py --model gpt-image-1.5 --background transparent --output-format webp\n\n# DALL-E 3 (note: count is automatically limited to 1)\npython3 {baseDir}/scripts/gen.py --model dall-e-3 --quality hd --size 1792x1024 --style vivid\npython3 {baseDir}/scripts/gen.py --model dall-e-3 --style natural --prompt \"serene mountain landscape\"\n\n# DALL-E 2\npython3 {baseDir}/scripts/gen.py --model dall-e-2 --size 512x512 --count 4\n```\n\n## Model-Specific Parameters\n\nDifferent models support different parameter values. The script automatically selects appropriate defaults based on the model.\n\n### Size\n\n- **GPT image models** (`gpt-image-1`, `gpt-image-1-mini`, `gpt-image-1.5`): `1024x1024`, `1536x1024` (landscape), `1024x1536` (portrait), or `auto`\n  - Default: `1024x1024`\n- **dall-e-3**: `1024x1024`, `1792x1024`, or `1024x1792`\n  - Default: `1024x1024`\n- **dall-e-2**: `256x256`, `512x512`, or `1024x1024`\n  - Default: `1024x1024`\n\n### Quality\n\n- **GPT image models**: `auto`, `high`, `medium`, or `low`\n  - Default: `high`\n- **dall-e-3**: `hd` or `standard`\n  - Default: `standard`\n- **dall-e-2**: `standard` only\n  - Default: `standard`\n\n### Other Notable Differences\n\n- **dall-e-3** only supports generating 1 image at a time (`n=1`). The script automatically limits count to 1 when using this model.\n- **GPT image models** support additional parameters:\n  - `--background`: `transparent`, `opaque`, or `auto` (default)\n  - `--output-format`: `png` (default), `jpeg`, or `webp`\n  - Note: `stream` and `moderation` are available via API but not yet implemented in this script\n- **dall-e-3** has a `--style` parameter: `vivid` (hyper-real, dramatic) or `natural` (more natural looking)\n\n## Output\n\n- `*.png`, `*.jpeg`, or `*.webp` images (output format depends on model + `--output-format`)\n- `prompts.json` (prompt \u2192 file mapping)\n- `index.html` (thumbnail gallery)\n\n## API key\n\n- `OPENAI_API_KEY` env var\n- Or set `skills.\\\"openai-image-gen\\\".apiKey` / `skills.\\\"openai-image-gen\\\".env.OPENAI_API_KEY` in `~/.CoWork-OSS/CoWork-OSS.json`\n",
   "parameters": [],
   "enabled": true
-}
+}

package/resources/skills/scripts/gen.py

@@ -0,0 +1,163 @@
+#!/usr/bin/env python3
+"""Minimal OpenAI image generation script (stdlib only)."""
+
+import argparse
+import base64
+import datetime
+import json
+import os
+import random
+import sys
+import urllib.error
+import urllib.request
+from pathlib import Path
+
+OPENAI_URL = "https://api.openai.com/v1/images/generations"
+
+DEFAULT_PROMPTS = [
+    "ultra-detailed studio photo of a lobster astronaut",
+    "cinematic wide shot of a glasshouse cafe in a tropical rainforest, morning mist",
+    "macro photograph of dewdrops on a spiderweb at sunrise, shallow depth of field",
+    "isometric cutaway of a cozy cabin library, warm lighting, rainy window",
+    "surreal desert landscape with floating rocks, golden hour light",
+]
+
+
+def _require_api_key() -> str:
+    api_key = os.environ.get("OPENAI_API_KEY")
+    if not api_key:
+        print("ERROR: OPENAI_API_KEY is not set", file=sys.stderr)
+        sys.exit(2)
+    return api_key
+
+
+def _post_json(payload: dict) -> dict:
+    api_key = _require_api_key()
+    data = json.dumps(payload).encode("utf-8")
+    req = urllib.request.Request(
+        OPENAI_URL,
+        data=data,
+        headers={
+            "Authorization": f"Bearer {api_key}",
+            "Content-Type": "application/json",
+        },
+        method="POST",
+    )
+    try:
+        with urllib.request.urlopen(req) as resp:
+            return json.load(resp)
+    except urllib.error.HTTPError as err:
+        body = err.read().decode("utf-8", "ignore")
+        print(f"OpenAI API error: {err.code} {err.reason}\n{body}", file=sys.stderr)
+        sys.exit(3)
+
+
+def _model_supports_gpt_image(model: str) -> bool:
+    return model.startswith("gpt-image")
+
+
+def build_payload(args: argparse.Namespace, prompt: str) -> dict:
+    payload = {
+        "model": args.model,
+        "prompt": prompt,
+        "n": args.count,
+        "size": args.size,
+        "response_format": "b64_json",
+    }
+
+    if args.quality:
+        payload["quality"] = args.quality
+    if args.style and args.model.startswith("dall-e-3"):
+        payload["style"] = args.style
+    if _model_supports_gpt_image(args.model):
+        if args.background:
+            payload["background"] = args.background
+        if args.output_format:
+            payload["output_format"] = args.output_format
+
+    return payload
+
+
+def _resolve_output_format(args: argparse.Namespace) -> str:
+    if args.output_format:
+        return args.output_format
+    return "png"
+
+
+def _write_index(out_dir: Path, files: list[str]) -> None:
+    index_path = out_dir / "index.html"
+    rows = "\n".join([f'<div><img src="{Path(f).name}" alt="image" /></div>' for f in files])
+    html = f"""<!doctype html>
+<html>
+<head>
+  <meta charset=\"utf-8\" />
+  <title>Image Gallery</title>
+  <style>body{{font-family:Arial,sans-serif}} img{{max-width:100%;height:auto;margin:8px 0}}</style>
+</head>
+<body>
+{rows}
+</body>
+</html>
+"""
+    index_path.write_text(html)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="OpenAI image generation script")
+    parser.add_argument("--prompt", help="Image prompt text")
+    parser.add_argument("--count", type=int, default=1)
+    parser.add_argument("--model", default="gpt-image-1")
+    parser.add_argument("--size", default="1024x1024")
+    parser.add_argument("--quality", default=None)
+    parser.add_argument("--out-dir", default="./out/images")
+    parser.add_argument("--output-format", default=None)
+    parser.add_argument("--background", default=None)
+    parser.add_argument("--style", default=None)
+    args = parser.parse_args()
+
+    if args.count < 1:
+        print("ERROR: --count must be >= 1", file=sys.stderr)
+        sys.exit(2)
+
+    if args.model.startswith("dall-e-3") and args.count != 1:
+        print("ERROR: dall-e-3 only supports --count 1", file=sys.stderr)
+        sys.exit(2)
+
+    prompt = args.prompt
+    if not prompt:
+        prompt = random.choice(DEFAULT_PROMPTS)
+
+    payload = build_payload(args, prompt)
+    result = _post_json(payload)
+    data = result.get("data", [])
+    if not data:
+        print("ERROR: No image data returned", file=sys.stderr)
+        sys.exit(4)
+
+    out_dir = Path(args.out_dir).expanduser()
+    out_dir.mkdir(parents=True, exist_ok=True)
+
+    ts = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
+    output_format = _resolve_output_format(args)
+
+    files: list[str] = []
+    for idx, item in enumerate(data, start=1):
+        b64 = item.get("b64_json")
+        if not b64:
+            continue
+        filename = out_dir / f"image_{ts}_{idx}.{output_format}"
+        with open(filename, "wb") as f:
+            f.write(base64.b64decode(b64))
+        files.append(str(filename))
+        print(f"MEDIA: {filename}")
+
+    # Write prompts.json and index.html for quick browsing
+    prompts_path = out_dir / "prompts.json"
+    prompts_path.write_text(json.dumps({"prompt": prompt, "files": files}, indent=2))
+    _write_index(out_dir, files)
+
+    print(f"Saved {len(files)} image(s) to {out_dir}")
+
+
+if __name__ == "__main__":
+    main()