agent-rs
What it is
agent-rs is a pure-Rust asynchronous runtime library for building LLM agents, featuring multi-provider support (Anthropic, OpenAI, Ollama) and end-to-end tool capabilities. It provides a unified event stream for text, tool usage, and cost tracking, along with structured permissions and Model Context Protocol (MCP) client support.
The problem it solves
Existing Rust LLM agent frameworks often lack unified multi-provider support, robust permission systems, or safe memory management, forcing developers to manually handle provider-specific quirks and security risks.
What you could use it for
- Building custom LLM-powered CLI tools in Rust
- Integrating autonomous agents into desktop applications or IDE plugins
- Creating secure agent backends with structured permission controls and cost monitoring
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.