What is Nelly?

Nelly is a comprehensive AI agent platform designed to empower users to build, test, distribute, and utilize AI agents through a unified interface. It facilitates the creation of custom AI agents using natural language, eliminating the need for coding expertise. Users can describe the desired functionality, and Nelly helps bring these intelligent agents to life.

The platform allows agents to be equipped with various tools, such as a browser and a database, to enhance their task-completion capabilities. Complex tasks can be managed by breaking them down into smaller problems and delegating them to specialized sub-agents, effectively creating a team of AI agents. Nelly also features a chat interface for natural conversations with these agents, providing real-time assistance and detailed activity tracking for transparency.

Features

Build with Studio: Create custom AI agents using natural language instructions, format instructions with headings and lists.
Toolkits for Agents: Equip agents with tools like a browser and database to help them accomplish tasks.
Testing and Refinement: Test AI agents with natural language and refine their behavior until they work as intended.
Sub-Agents for Complex Tasks: Break down complex problems and delegate them to specialized sub-agents to build a team of agents.
Natural Language Chat: Interact with AI agents through natural, flowing conversations without needing special commands or syntax.
Threaded Conversations: Keep conversations organized in threads for better organization and improved performance.
Agent Departments: Organize AI agents into departments using a drag-and-drop interface.
Detailed Activity Tracking: Get insights into agents' activities, including tools they use, browser screenshots, and step-by-step progress.
Custom LLM Providers: Connect personal AI model providers such as Anthropic, OpenAI, OpenRouter, and LM Studio for enhanced control (Paid Feature).
Marketplace (Coming Soon): A future feature to share and monetize created AI agents and tools.

Use Cases

Building custom AI agents for specific tasks without requiring coding skills.
Automating simple to complex workflows by assembling a team of specialized AI agents.
Interacting with AI agents via natural conversation for real-time assistance and task execution.
Organizing and managing multiple AI agents efficiently for various functions within an organization or project.
Testing, iterating, and refining AI agent behavior to achieve optimal performance and desired outcomes.
Extending AI agent capabilities by integrating custom tools and preferred Large Language Model (LLM) providers.
Sharing and potentially monetizing user-created AI agents and tools on a dedicated marketplace (future capability).

FAQs

Where do my AI agents run?

Your AI agents and their tools run locally on your computer, while the LLM that powers them operates in the cloud.
How is my data handled?

All your AI agents and their data are stored locally on your computer. Nelly only stores your user account information.
What information is sent to the cloud when using Nelly?

Nelly only transmits your conversation history and context to the LLM provider. This data is not stored by Nelly.
Which LLM does Nelly use by default?

Nelly currently uses Anthropic's Claude 3.7 Sonnet, though this may change based on performance, cost, and other factors.
How does Nelly handle website interactions with its browser tool?

Nelly agents operate at human-like speeds and use only one browser at a time to ensure compliance with website terms of service.