# Connection Providers

This page lists all available connection providers in AgenticFlow with their exact configuration fields and credential sources.

***

## MCP (Model Context Protocol) Connections

### MCP Connection

* **Category**: `mcp`
* **Description**: Universal connection to MCP servers providing tools, prompts, and resources
* **Authentication Types**: None, OAuth 2.0 (auto-discovery)
* **Configuration Fields**:
  * **url** (required) - MCP server endpoint URL
  * **transport** - `sse` (Server-Sent Events) or `streamable_http`
  * **headers** (optional) - Custom HTTP headers as key-value pairs
  * **auth\_type** - `none` or `oauth`
  * **client\_metadata** (optional):
    * **image\_url** - Icon/logo URL for the connection
    * **docs\_url** - Documentation URL
* **Documentation**: [MCP Specification](https://modelcontextprotocol.io/)
* **Learn More**: [What are MCPs?](/integrations/overview.md)

***

## AI Model Providers

### OpenAI

* **Category**: `openai`
* **Description**: Access GPT models and OpenAI services (ChatGPT, GPT-4, DALL-E, Whisper)
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your OpenAI API key
  * **base\_url** (optional) - Custom base URL (default: `https://api.openai.com/v1`)
* **Get Credentials**: [OpenAI API Keys](https://platform.openai.com/api-keys)
* **Documentation**: [OpenAI API Reference](https://platform.openai.com/docs/api-reference)

### Claude (Anthropic)

* **Category**: `claude`
* **Description**: Access Claude AI models from Anthropic (Claude 3.5 Sonnet, Claude 3 Opus)
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your Anthropic API key
* **Get Credentials**: [Anthropic Console](https://console.anthropic.com/settings/keys)
* **Documentation**: [Claude API Docs](https://docs.anthropic.com/en/api/getting-started)

### Google Gemini

* **Category**: `google_gen_ai`
* **Description**: Access Google's Gemini AI models (Gemini Pro, Gemini Flash)
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your Google AI Studio API key
* **Get Credentials**: [Google AI Studio](https://aistudio.google.com/app/apikey)
* **Documentation**: [Gemini API Docs](https://ai.google.dev/gemini-api/docs)

### DeepSeek

* **Category**: `deepseek`
* **Description**: Access DeepSeek AI models (DeepSeek V4 Pro, DeepSeek V4 Flash)
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your DeepSeek API key
  * **base\_url** - API base URL (default: `https://api.deepseek.com`)
* **Get Credentials**: [DeepSeek Platform](https://platform.deepseek.com/api_keys)
* **Documentation**: [DeepSeek API Docs](https://api-docs.deepseek.com/)

### Groq

* **Category**: `groq`
* **Description**: High-speed LLM inference with Groq's LPU technology (Llama, Mixtral, Gemma)
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your Groq API key from console
  * **base\_url** - API base URL (default: `https://api.groq.com`)
* **Get Credentials**: [Groq Console](https://console.groq.com/keys)
* **Documentation**: [Groq API Docs](https://console.groq.com/docs/quickstart)

### Perplexity

* **Category**: `perplexity`
* **Description**: Search-augmented AI models from Perplexity
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your Perplexity API key (starts with `pplx-`)
* **Get Credentials**: [Perplexity API Settings](https://www.perplexity.ai/settings/api)
* **Documentation**: [Perplexity API Docs](https://docs.perplexity.ai/)
* **Note**: Requires credit balance to generate API keys

***

## AI Infrastructure & Services

### Replicate

* **Category**: `replicate`
* **Description**: Run open-source AI models in the cloud (FLUX, Stable Diffusion, etc.)
* **Authentication**: API Token
* **Configuration Fields**:
  * **api\_token** (required) - Your Replicate API token (starts with `r8_`)
* **Get Credentials**: [Replicate API Tokens](https://replicate.com/account/api-tokens)
* **Documentation**: [Replicate Docs](https://replicate.com/docs)

### FAL.ai

* **Category**: `fal`
* **Description**: Fast AI inference platform for generative models
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your FAL.ai API key
* **Get Credentials**: [FAL.ai Dashboard](https://fal.ai/dashboard/keys)
* **Documentation**: [FAL.ai Docs](https://docs.fal.ai/)

### Straico

* **Category**: `straico`
* **Description**: Multi-model AI platform with unified API
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your Straico API key
* **Get Credentials**: [Straico Dashboard](https://straico.com/) → Profile → API Settings
* **Documentation**: [Straico API Reference](https://documenter.getpostman.com/view/5900072/2s9YyzddrR)

### PixelML

* **Category**: `pixelml`
* **Description**: Enterprise AI platform for custom models
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your PixelML API key
  * **base\_url** (optional) - Custom base URL for self-hosted deployments
* **Get Credentials**: [PixelML Platform](https://platform.pixelml.com/)

***

## Specialized Tools & Services

### Firecrawl

* **Category**: `firecrawl`
* **Description**: Web scraping and crawling API optimized for AI/LLMs
* **Authentication**: API Key
* **Configuration Fields**:
  * **connections** (list of connections):
    * **api\_key** (required) - Your Firecrawl API key (starts with `fc-`)
    * **base\_url** - API base URL (default: `https://api.firecrawl.dev`)
* **Get Credentials**: [Firecrawl Dashboard](https://www.firecrawl.dev/) → Login → View API Key
* **Documentation**: [Firecrawl API Docs](https://docs.firecrawl.dev/api-reference/introduction)

### Tavily

* **Category**: `tavily`
* **Description**: Search API optimized for AI agents and LLMs
* **Authentication**: API Key
* **Configuration Fields**:
  * **api\_key** (required) - Your Tavily API key (starts with `tvly-`)
* **Get Credentials**: [Tavily Dashboard](https://app.tavily.com/home)
* **Documentation**: [Tavily Docs](https://docs.tavily.com/)
* **Free Tier**: 1,000 API credits per month

### Telegram

* **Category**: `telegram`
* **Description**: Telegram Bot API for messaging and automation
* **Authentication**: Bot Token
* **Configuration Fields**:
  * **token** (required) - Your Telegram Bot token from BotFather
* **Get Credentials**:
  1. Open Telegram and search for `@BotFather`
  2. Send `/newbot` command
  3. Follow prompts to create your bot
  4. Copy the provided token
* **Documentation**: [Telegram Bot API](https://core.telegram.org/bots/api)

***

## OAuth 2.0 Connections

### YouTube

* **Category**: `youtube`
* **Description**: YouTube Data API v3 for video and channel management
* **Authentication**: OAuth 2.0
* **Configuration Fields**:
  * **auth\_url** - `https://accounts.google.com/o/oauth2/auth`
  * **token\_url** - `https://oauth2.googleapis.com/token`
  * **scope** - `https://www.googleapis.com/auth/youtube.upload`
  * **client\_id** - Configured by platform
  * **client\_secret** - Configured by platform
* **OAuth Flow**: Handled automatically by AgenticFlow
* **Documentation**: [YouTube Data API](https://developers.google.com/youtube/v3)

## Related Documentation

* [What are MCPs?](/integrations/overview.md)
* [Using Connections in Agents](/ai-agents/03-agents.md)
* [Using Connections in Workflows](/workflows/04-workflows.md)
* [Security Overview](/policies/security-overview.md)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.agenticflow.ai/integrations/connection-providers.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.