CrewAI
Multi-agent automation framework & studio to build and run AI crews
Open-Source Projects
Discover top open-source software, updated regularly with real-world adoption signals.
Discover top open-source software, updated regularly with real-world adoption signals.
TypeScript framework for production-ready AI agents with observability
VoltAgent lets developers build, orchestrate, and monitor AI agents using TypeScript, offering memory, workflow orchestration, tool integration, multi-agent supervision, and built-in LLM observability out of the box.
VoltAgent is a TypeScript‑first platform that enables teams to create intelligent agents ready for production from day one. It bundles memory adapters, tool registries, and a declarative workflow engine so developers can focus on agent logic rather than infrastructure.
The core runtime supports typed agents, supervisors for coordinating sub‑agents, and provider‑agnostic LLM integration (OpenAI, Anthropic, Google, etc.). Workflows are defined declaratively, allowing human‑in‑the‑loop steps, suspension, and resumption. Built‑in VoltOps provides real‑time execution traces, performance metrics, and OpenTelemetry‑compatible dashboards without external services.
Agents run in any Node.js environment, from local servers to serverless platforms. A CLI (npm create voltagent-app@latest) scaffolds projects with example agents, tools, and workflows, and the optional MCP docs server lets LLMs access VoltAgent documentation directly. Monitoring is accessible via the VoltOps console at console.voltagent.dev.
When teams consider VoltAgent, these hosted platforms usually appear on the same shortlist.
Looking for a hosted option? These are the services engineering teams benchmark against before choosing open source.
Expense approval automation
Automates expense report validation, routes high‑value requests to managers, and records decisions with audit trails.
AI‑powered coding assistant
Provides contextual code suggestions by retrieving documentation via MCP server and executing tool calls.
Customer support triage
Routes inquiries to specialized sub‑agents, uses memory to retain conversation context, and logs interactions for performance analysis.
Multi‑agent research assistant
Coordinates a team of agents to gather data, summarize findings, and present a cohesive report with real‑time monitoring.
Run `npm create voltagent-app@latest` which scaffolds a TypeScript project with example agents and workflows.
Yes, configure a different provider (OpenAI, Anthropic, Google, etc.) in the model config; the framework abstracts the provider.
Real‑time execution traces, tool usage logs, performance metrics, and OpenTelemetry‑compatible dashboards accessible via the VoltOps console.
Agents can attach durable memory adapters (e.g., LibSQL) to persist context; the default is an in‑memory store.
The core runtime can run in any Node.js environment, including serverless platforms, as long as required dependencies are available.
Project at a glance
ActiveLast synced 4 days ago
