text stringlengths 0 59.1k |
|---|
- `lib/agent.ts`: VoltAgent `Agent` plus `getCurrentTime` tool |
- `lib/bot.ts`: Chat SDK adapter setup, mention/subscribed/action handlers |
- `app/api/webhooks/[platform]/route.ts`: dynamic webhook route with `waitUntil` |
If Slack credentials are missing, the webhook route returns a clear runtime error instead of failing the build. |
<|endoftext|> |
# source: VoltAgent__voltagent/website/recipes/ad-creator.md type: docs |
--- |
id: ad-creator |
slug: ai-instagram-ad-agent |
title: AI Ads Generator Agent |
description: Build Instagram ads with VoltAgent, BrowserBase and Gemini. |
repository: https://github.com/VoltAgent/voltagent/tree/main/examples/with-ad-creator |
--- |
This example implements an Instagram ad generator with VoltAgent. The system uses BrowserBase Stagehand to analyze landing pages, extracts brand data, and generates visuals through Google Gemini AI. |
The implementation includes: |
- Web page navigation and analysis via BrowserBase Stagehand |
- Brand data extraction (tone, audience, value propositions) |
- Screenshot capture for visual reference |
- Creative brief generation through AI models |
- Ad visual generation using Google Gemini |
- Multi-agent workflow orchestration |
### Setup |
<Info title="Required accounts and API keys:"> |
- Sign in to [VoltOps LLM Observability platform](https://console.voltagent.dev/login) |
- A [BrowserBase account](https://browserbase.com) with API key and project ID |
- Google Generative AI access for [Gemini API](https://ai.google.dev/gemini-api) |
- An OpenAI API key from [platform.openai.com](https://platform.openai.com/api-keys) |
</Info> |
#### Get the example code |
Clone the ad generator example using the VoltAgent CLI: |
```bash |
npm create voltagent-app@latest -- --example with-ad-creator |
cd with-ad-creator |
``` |
You can find the source code of this example [here](https://github.com/VoltAgent/voltagent/tree/main/examples/with-ad-creator). |
#### Configure environment variables |
Create a `.env` file with your API keys: |
```env |
# Language models |
GOOGLE_GENERATIVE_AI_API_KEY=your_google_generative_ai_api_key |
OPENAI_API_KEY=your_openai_api_key |
# BrowserBase Stagehand |
BROWSERBASE_API_KEY=your_browserbase_api_key |
BROWSERBASE_PROJECT_ID=your_browserbase_project_id |
# Optional VoltOps tracing |
VOLTAGENT_PUBLIC_KEY=your_public_key |
VOLTAGENT_SECRET_KEY=your_secret_key |
``` |
#### Set Up BrowserBase |
Configure BrowserBase: |
1. Create a BrowserBase account at [browserbase.com](https://browserbase.com) |
2. Navigate to your dashboard and create a new project |
3. Copy your API key from the API Keys section |
4. Copy your Project ID from the project settings |
5. Add both values to your `.env` file |
For BrowserBase configuration details, see [BrowserBase documentation](https://docs.browserbase.com/introduction). |
 |
#### Configure Google Gemini API |
Set up Google Gemini for image generation: |
1. Go to [Google AI Studio](https://aistudio.google.com/apikey) |
2. Click "Get API Key" and create a new API key |
3. Enable the Generative AI API in your Google Cloud Console if needed |
4. Add the API key to your `.env` file as `GOOGLE_GENERATIVE_AI_API_KEY` |
Google Gemini handles creative brief generation and image creation. |
#### Start the Development Server |
```bash |
npm run dev |
``` |
After server startup: |
```bash |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.