> ## Documentation Index
> Fetch the complete documentation index at: https://docs.enconvo.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Built-in Extensions
> Complete guide to all built-in extensions in Enconvo
## Overview
EnConvo ships with 80+ built-in extensions organized into functional categories. Each extension provides specialized capabilities that can be used standalone, combined in workflows, or called by AI agents as tools.
All built-in extensions are open source and available on [GitHub](https://github.com/enconvo). You can customize or fork any extension to fit your needs.
## AI & Chat
Core AI interaction extensions that power EnConvo's conversational interface.
| Extension | Description | Key Features |
| -------------------------------------- | -------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------- |
| **AI (chat\_with\_ai)** | Multi-modal AI chat agent with tool use and knowledge base retrieval | Agent mode with planning, multi-LLM support, dynamic context awareness, customizable system prompts, post-actions |
| **Agent (agent)** | AI agent management with the Mavis personal assistant | Create, list, and message agents; configurable prompts, tools, and session handling; built-in Mavis assistant |
| **Custom Bot (custom\_bot)** | Registry for user-created custom AI bots and agents | Dynamic command creation, personalized bot configurations |
| **Writing Toolkit (writing\_package)** | AI-powered writing commands | Fix spelling/grammar, improve writing, summarize, explain, rewrite, tone adjustment, proofreading, dictation polishing |
| **Web Search (internet\_browsing)** | Real-time web search with AI synthesis | Configurable search providers, LLM-synthesized cited answers |
The **AI** extension is the primary chat interface. Use `@` in SmartBar to select tools, and `#` to add context sources like screenshots, browser tabs, or clipboard content.
## Language Models
Extensions for managing and connecting to AI model providers.
| Extension | Description | Key Features |
| ----------------------------------------------- | -------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| **AI Model Providers (llm)** | Unified LLM provider integrations for 20+ services | OpenAI, Anthropic, Google Gemini, Ollama, Groq, Mistral, DeepSeek, X.AI, Azure OpenAI, LM Studio, and more; streaming chat completion; model management APIs |
| **Embedding Providers (embeddings\_providers)** | Text embedding for semantic search and knowledge base indexing | OpenAI, Voyage AI, Ollama, SiliconFlow, Enconvo Cloud; configurable concurrency and model selection |
| **Reranker Providers (reranker\_providers)** | Document reranking for improved search relevance | Voyage AI, SiliconFlow, Enconvo Cloud reranking models |
| **MLX (mlx\_manage)** | Apple Silicon local AI model management | Local LLM chat, vision-language models, TTS, STT via MLX; Hugging Face model downloads |
Use the **Enconvo Cloud Plan** provider for zero-configuration access to multiple models. For privacy-sensitive work, configure **Ollama** or **MLX** for fully local inference.
## Voice & Audio
Text-to-speech, speech recognition, transcription, and audio utilities.
| Extension | Description | Key Features |
| ------------------------------------------------------------ | -------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- |
| **TTS (tts)** | Text-to-speech commands | Read aloud, text-to-audio file conversion, sound effects, SRT-to-audio, Gemini single/multi-speaker TTS |
| **TTS Providers (tts\_providers)** | 11+ TTS provider backends | OpenAI, Microsoft Azure, ElevenLabs, MiniMax, Speechify, Google Cloud, xAI, macOS system, Edge TTS; voice listing and speed configuration |
| **Dictation Model Providers (speech\_recognize\_providers)** | 15+ real-time speech-to-text providers | Groq Whisper, OpenAI, Deepgram, AssemblyAI, ElevenLabs, Google Gemini, Microsoft Azure, NVIDIA Parakeet, Qwen ASR, local Whisper |
| **Transcription Providers (transcription\_providers)** | File-based transcription for long-form audio | Same provider backends as dictation; optimized for longer recordings |
| **Audio Utils (audio\_utils)** | Audio file compression | ffmpeg-based compression with configurable quality; batch processing of MP3, WAV, M4A, AAC, OGG, FLAC |
| **My Recordings (my\_recordings)** | Recording management | Organize and access voice recordings |
For the best free dictation experience, use **Microsoft (Enconvo Cloud Plan) Free** which requires no API key.
## Content & Knowledge
Extensions for managing knowledge, searching, reading web content, and crawling websites.
| Extension | Description | Key Features |
| ----------------------------------------------------------- | --------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Knowledge Base (knowledge\_base)** | Document management and semantic search | Create knowledge bases, ingest files/text/webpages/audio/video, vector similarity search via LanceDB, personal memory system with automatic knowledge extraction |
| **Search (search)** | Unified search engine | Search commands, files, applications, skills, and conversations; fuzzy matching, context-aware recommendations, real-time index updates |
| **Web Fetch (link\_reader)** | Fetch and extract content from URLs | Convert webpages to markdown/text, summarize pages using LLM |
| **Link Reader Providers (link\_reader\_providers)** | Web scraping backends | Local Fetch, Crawl4AI, Jina AI, Firecrawl, Enconvo Cloud |
| **Website Crawler (website\_crawler)** | Multi-page website crawling | Configurable depth and page limits, output as text or saved to file |
| **Website Crawler Providers (website\_crawler\_providers)** | Crawler backends | Crawl4AI (local, free), Firecrawl (cloud API) |
| **Web Search Providers (browsing\_providers)** | Web search backends for AI agents | Tavily, Bing, Google Custom Search, Exa, Gemini grounding |
| **Document Loader Providers (document\_loader\_providers)** | Document text extraction | Mistral OCR for structured text and markdown from documents and images |
Combine **Knowledge Base** with **Embedding Providers** and **Reranker Providers** for a powerful RAG (Retrieval-Augmented Generation) pipeline. Ingest your documents, then ask questions in chat with knowledge base context enabled.
## Image & Video
Image generation, video creation, compression, OCR, and media utilities.
| Extension | Description | Key Features |
| ----------------------------------------------------------- | ------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------ |
| **Image Create (image\_generation)** | Generate and edit images from text prompts | Background removal, multiple output formats, batch generation |
| **Image Generate Providers (image\_generation\_providers)** | 7+ image generation backends | OpenAI DALL-E, Google Gemini, Fal.ai, Azure, Together AI, Straico; text-to-image and image editing |
| **Video Generation (video\_generation)** | AI video creation | Text-to-video, image-to-video using Sora, Kling, Hailuo, Veo, Wan |
| **Video Generate Providers (video\_generation\_providers)** | Video generation backends | Enconvo Cloud (fal.ai), OpenAI Sora, Fal.ai direct |
| **Image Compress (compress\_image)** | Image compression | Caesium engine, adjustable quality, batch compression, overwrite or output to folder |
| **Video Utils (video\_utils)** | Video utilities | Compress video via ffmpeg, extract audio to MP3, download videos from YouTube/TikTok/Instagram/Reddit/Twitter via yt-dlp |
| **OCR (ocr\_action)** | Text extraction from images | Screenshot OCR, silent clipboard copy, interactive SmartBar display |
| **OCR Providers (ocr\_providers)** | OCR backends | Apple native Vision OCR, AI model-based OCR via LLM vision, Mistral OCR |
## Productivity
Workflow automation, scheduling, prompt management, and utility extensions.
| Extension | Description | Key Features |
| ------------------------------------ | --------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Workflow (workflow)** | Visual workflow automation engine | Node-based execution, manual/scheduled/file-change triggers, conditional branching, loop iteration, AI parameter generation, Jinja2 templating, run logging |
| **Cron (cron)** | Scheduled task management | Create, list, update, delete, toggle cron jobs; execute commands on recurring schedules; execution logs |
| **Prompt (prompt)** | Reusable prompt template system | Create, manage, execute prompt templates with variable substitution, diff highlighting, post-actions |
| **Variables (variables)** | System variable providers | Active app info, screenshots, clipboard, selected text, Finder paths, browser tabs, date/time for use in workflows and templates |
| **Quick Links (quick\_links)** | Parameterized URL launcher | Open URLs in configurable browser; presets for Google, YouTube, Twitter/X search |
| **Skills Manager (skills\_manager)** | Skill installation and updates | Install from Enconvo Store, GitHub, or ZIP; version checking; auto-install built-in skills |
| **Enconvo (enconvo)** | Core platform APIs | 30+ internal endpoints for extension management, command configuration, credential loading, cache operations, user profile |
## Communication
Email, messaging, and channel integrations.
| Extension | Description | Key Features |
| -------------------------------------- | ------------------------------ | --------------------------------------------------------------------------------------------- |
| **Gmail (gmail)** | Gmail API integration | Send, draft, read, search, delete emails; manage labels; batch operations |
| **IM Channels (im\_channels)** | Messaging platform integration | Slack, Telegram, Discord, Feishu/Lark; send and read messages; channel listener service |
| **Message Manager (message\_manager)** | Chat message persistence | SQLite + Drizzle ORM; store, query, paginate conversation messages across agents and sessions |
Set up **IM Channels** to receive AI agent notifications directly in Slack or Telegram. Combine with **Cron** to schedule automated reports sent to your team channels.
## Apple Integration
Native macOS application integrations using AppleScript and system APIs.
| Extension | Description | Key Features |
| -------------------------------------- | --------------------------- | -------------------------------------------------------------------------------------------------------- |
| **Apple Mail (apple\_mail)** | Apple Mail integration | Compose, read, manage emails; archive, trash, delete, move between mailboxes; toggle read status |
| **Apple Reminders (apple\_reminders)** | Apple Reminders integration | Create, update, delete, list reminders; due dates, priorities, recurrence rules, location-based triggers |
| **Shortcuts (apple\_shortcuts)** | Apple Shortcuts integration | Dynamically populated commands based on shortcuts available on the system |
| **Apple Calendar (calender)** | Apple Calendar integration | Full CRUD for events and calendars; recurrence rules, alarms, all-day events |
These extensions work through native macOS APIs, so they require appropriate permissions. Grant access in System Settings when prompted.
## System & Control
System utilities, file operations, window management, and computer control.
| Extension | Description | Key Features |
| ---------------------------------------- | ------------------------------- | ------------------------------------------------------------------------------------------------------ |
| **Mac Applications (application)** | AI-powered app assistants | Context-aware prompts for 12+ apps: Finder, Calendar, Mail, Notes, Obsidian, Microsoft Office |
| **File System (file\_system)** | File system tools for AI agents | Read, write, edit files; search with ripgrep (grep); find files by glob patterns |
| **Window (window)** | Window management | PopBar configuration, Mini SmartBar setup, display mode options |
| **System Tools (system)** | macOS system utilities | Toggle dark mode, hidden files, mute, full screen; web search; URL opening |
| **Computer Control (computer\_control)** | Accessibility API control | Navigate UI trees, click elements, type text, screenshots, scroll; works with any accessible macOS app |
## Development & Automation
Code execution, browser automation, and protocol integrations.
| Extension | Description | Key Features |
| ------------------------------ | ----------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Code Runner (code\_runner)** | Shell command execution | Bash/zsh commands with auto Python venv activation, configurable working directory, timeout handling |
| **MCP (mcp)** | Model Context Protocol client | Connection pooling, stdio/HTTP/SSE transports, OAuth authentication for MCP server tools |
| **Browser Use (browser-use)** | Browser automation | Navigate, click, type, screenshots, read page content, manage tabs; works with Chrome, Edge, Brave, Arc, Vivaldi, Opera, Firefox via Companion extension |
**Browser Control** uses your existing browser sessions -- all your login cookies and extensions work automatically. Install the Enconvo Companion browser extension for the best experience.
## Translation
Multi-language translation with multiple provider backends.
| Extension | Description | Key Features |
| ---------------------------------------------- | ------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- |
| **Translate (translate)** | Text and word translation | AI-powered LLM, DeepL, Google Translate, Youdao, Caiyun; auto language detection, dictionary mode, screenshot OCR translation, TTS playback |
| **Translate Providers (translate\_providers)** | Translation backends | LLM-based, DeepL, Google Translate, Youdao, Caiyun, Enconvo Cloud |
## Other Utilities
Additional tools for export, credentials, YouTube, and more.
| Extension | Description | Key Features |
| ------------------------------------ | ---------------------------------- | ------------------------------------------------------------------------------------------------------------------ |
| **Exporter (exporter)** | Export conversations and documents | Save as Markdown, PDF, or audio (MP3) via TTS |
| **Credentials (credentials)** | Centralized API key management | 40+ AI service providers; key validation; secure storage in macOS Keychain |
| **YouTube (youtube)** | YouTube tools | Extract video transcripts with timestamps and language selection; download videos/audio via yt-dlp |
| **EnConvo WebApp (enconvo\_webapp)** | Management UI | Extension store, agent creation, workflow builder (React Flow), knowledge base management, MCP server installation |
## Extension Architecture
### How Extensions Work
Every extension follows a consistent pattern:
1. **Package Definition**: `package.json` declares commands, preferences, and metadata
2. **Command Handlers**: TypeScript files in `src/` export entry functions
3. **API Routes**: Files in `src/api/` are auto-discovered as HTTP-like endpoints
4. **Provider Pattern**: Extensions with `commandType: "provider"` register capabilities that consumer extensions select
### Command Key Format
All extensions use the format `extensionName|commandName` for targeting. For example:
* `chat_with_ai|chat` -- the main chat command
* `llm|open_ai` -- OpenAI LLM provider
* `gmail|send_email` -- Gmail send tool
### Provider Selection
Many extensions use a provider pattern where you choose which backend to use:
```
chat_with_ai (consumer) → llm|open_ai (provider)
tts (consumer) → tts_providers|openai_tts_provider (provider)
translate (consumer) → translate_providers|deepl_translate_provider (provider)
```
Configure default providers in each extension's settings, or let AI agents choose automatically.
## Installing Additional Extensions
Beyond built-in extensions, you can install community extensions:
Settings → Extensions → Browse, or use the `/` command in SmartBar
Find extensions by name or category
Click Install -- the extension is downloaded and available immediately
Some extensions require API keys or preferences. Configure in Settings → Extensions → \[Extension Name]
## Next Steps
Create your own custom extensions
Complete API reference for extension development
Combine extensions into automated workflows
Extend AI capabilities with task-specific skills