Open Source AI Coding Agents - Page 3

Letta Code is a memory-first CLI coding agent built on the Letta platform that offers developers a persistent AI assistant capable of learning and improving over time rather than resetting state each session, giving agents a sense of continuity and context across coding tasks. Unlike traditional session-based coding tools, Letta Code attaches a long-lived agent to a working directory so that the agent accumulates memory about a project’s structure, preferences, and history, effectively acting as a collaborative partner rather than a stateless helper. Users can initialize and connect the agent to various models, including popular large language models, and issue commands, refactor code, or ask context-aware questions directly in the terminal, with memory retained across multiple interactions.

Open Vibe is an open-source course and agent workflow that turns Claude Code, Codex, Copilot, Open Code, or another terminal-capable AI coding agent into a SaaS-building assistant. It is built around Open SaaS, a free Wasp-powered SaaS boilerplate, so learners can create a real app while understanding the architecture behind production-ready SaaS systems. The workflow starts with setup instructions that install Node.js, install the Wasp CLI, and verify the local environment. After creating a new Wasp app, the user opens an AI coding agent inside the project and lets it fetch course module instructions. The agent then works as both tutor and pair programmer, explaining the system while helping the user build features from plain-language requests. Progress is tracked through JSON files written into the project, making the learning path structured while still letting the user build their own app idea.

A terminal coding agent that reads and edits your project, runs shell commands, uses Git and the web, and analyzes logs without burning your model tokens. 29 built-in tools, direct-command execution, persistent chat sessions, and multi-provider LLM support.

mac code is a local AI coding agent designed to run large language models directly on Apple Silicon machines without relying on cloud services, effectively transforming a Mac into a self-contained AI development environment. The project focuses on enabling models that traditionally exceed available RAM to run efficiently by streaming model weights from SSD storage, thereby overcoming hardware limitations through innovative memory management techniques. It operates as a CLI-based assistant that routes user prompts into different execution paths such as chat, shell commands, or web search, functioning as a multi-purpose development agent. The system integrates with inference engines like llama.cpp and Apple’s MLX framework, allowing users to run models up to 35B parameters locally with varying performance trade-offs.

nanocode is a minimalist coding agent implementation designed as a compact alternative to Claude Code, packaged in a single Python file with no external dependencies and totaling around 250 lines of code. It implements a full agentic loop where the model can reason, decide when to use tools, execute those tools, and iterate until producing a final answer, making it useful for simple AI-assisted coding workflows. It includes a set of integrated tools such as read, write, edit, glob, grep, and bash that let the agent interact with the file system and shell commands directly from the terminal, and it keeps a conversation history with colored terminal output for readability. The project exemplifies how lightweight architectures can still support practical agent workflows without complex infrastructure, making it suitable for developers exploring agent frameworks or building custom coding assistants.