VoltAgent – Open-Source TypeScript AI Agent Framework

4 months ago 2

Decorative dot pattern background

Open Source

Escape no-code limits and scratch-built overhead. Build, customize, and orchestrate AI agents with full control, speed, and a great DevEx.

import { VoltAgent, Agent, createTool, createHooks } from "@voltagent/core"; import { VercelAIProvider } from "@voltagent/vercel-ai"; import { openai } from "@ai-sdk/openai"; import { fetchRepoContributorsTool } from "./tools"; import { fetchRepoStarsTool } from "./tools"; const starsFetcherAgent = new Agent({ name: "Stars Fetcher", description: "Fetches the number of stars for a GitHub repository using the GitHub API", llm: new VercelAIProvider(), model: openai("gpt-4o-mini"), tools: [fetchRepoStarsTool], }); const contributorsFetcherAgent = new Agent({ name: "Contributors Fetcher", description: "Fetches the list of contributors for a GitHub repository using the GitHub API", llm: new VercelAIProvider(), model: openai("gpt-4o-mini"), tools: [fetchRepoContributorsTool], }); const analyzerAgent = new Agent({ name: "Repo Analyzer", description: "Analyzes repository statistics and provides insights", llm: new VercelAIProvider(), model: openai("gpt-4o-mini"), }); const supervisorAgent = new Agent({ name: "Supervisor", description: `You are a GitHub repository analyzer. When given a GitHub repository URL or owner/repo format, you will: 1. Use the StarsFetcher agent to get the repository's star count 2. Use the ContributorsFetcher agent to get the repository's contributors 3. Use the RepoAnalyzer agent to analyze this data and provide insights Example input: https://github.com/voltagent/voltagent or voltagent/voltagent `, llm: new VercelAIProvider(), model: openai("gpt-4o-mini"), subAgents: [starsFetcherAgent, contributorsFetcherAgent, analyzerAgent], }); new VoltAgent({ agents: { supervisorAgent, }, });

Read-only

Enterprise-level AI agents

Complete toolkit for enterprise level AI agents

Design production-ready agents with unified APIs, tools, and memory.

Enable agents to invoke functions, interact with systems, and perform actions.

Seamlessly switch between different AI providers with a simple code update.

Experiment, fine-tune, and iterate your AI prompts in an integrated environment.

Store and recall interactions to enhance your agents intelligence and context.

INTEGRATIONS

Easily connect with 40+ apps in no time

Integrate your AI agents with your preferred tools and services effortlessly.

Observability

Stay in control at every stage

From tracking deployments to debugging and live interaction, VoltAgent gives you full visibility into your AI agents.

Deployment

Deploy your Agents in seconds with VoltAgent Deployment.

Initializing build environment

Deploying to VoltAgent's global network

10:30:01

Cloning repository...

10:30:02

From https://github.com/your-org/your-app

10:30:02

* branch main -> FETCH_HEAD

10:30:02

HEAD is now at a1b2c3d feat: add new feature

10:30:02

Success: Finished cloning repository files

10:30:03

Using Node.js v20.19.0

10:30:03

Running `npm install`...

10:30:28

Installed 1450 packages in 25.3s

10:30:29

Running `npm run build`...

10:30:29

> tsc && node scripts/postbuild.js

10:30:35

Compiling TypeScript files...

10:30:36

[VoltAgent] Validating agent schemas...

10:30:38

Build successful! Artifacts generated.

10:30:38

[VoltAgent] Starting deployment process...

10:30:40

Uploading build artifacts (2.1MB)...

10:30:45

Upload complete. Verifying deployment...

10:30:46

[VoltAgent] Deployed! 'my-agent-v1.2' is live.

Debugging

Debug and analyze your VoltAgent-powered AI agent's behavior with visual flows.

Observability

Connect your VoltAgent-powered AI agents to popular observability platforms.

AI Chat

Interact with your AI agent through natural language chat interface.

Tools

Social Analysis, Trend Detection, Content Calendar, Performance Metrics, Competitor Analysis

Instructions

You are a marketing assistant specialized in campaign analysis, content strategy, and social media optimization. Analyze data to provide actionable insights and automate routine marketing tasks.

Agent, analyze last month's social media campaign performance and provide recommendations for improvement.

Analysis complete. Last month's campaign reached 45% more users but had 12% lower conversion rate compared to previous campaigns. Key findings: 1) Video content performed 3x better than static images. 2) Posts published between 6-8pm had highest engagement. 3) Product demonstration posts generated most conversions. Recommendation: Increase video content by 40%, focus on product demonstrations, and schedule more posts during evening hours.

Create a content calendar for next month based on these insights. Include optimal posting times and content types.

Content calendar created. I've scheduled 15 posts across platforms with 60% video content focused on product demos. Primary posting times are Tuesdays and Thursdays 6-8pm, with additional posts Monday and Friday mornings based on secondary engagement peaks. I've included hooks for upcoming product launch and integrated seasonal marketing themes. Calendar has been added to your Marketing Projects workspace and synced with team collaboration tools.

Blog

All about AI Agents

We are sharing our knowledge about AI Agents.

Community

Join the movement

Our growing open source community building the future of AI agents.

Read Entire Article