Code agents

Simple tool for formatting code for LLMs.

Dynamically generate JSON Schema from Python code.

`LLMfy` is a framework for developing applications with large language models (LLMs).

Sync LLM agent configuration files (AGENTS.md, skills/) across different coding assistants

Parse and match .llmignore files — the .gitignore for AI. Supports .cursorignore, .aiignore, .aiexclude, .claudeignore, .codeiumignore, .geminiignore, .aiderignore, .clineignore, .rooignore, .augmentignore, .copilotignore, .repomixignore.

Isolated Python code execution with resource limits

No-code pipeline automation — visual DAG builder with 112+ blocks

AI-native frontend development where AI builds, inspects, and operates real apps, dashboards, and tools

A code obfuscator to confuse LLMs

Making agents cuter

CLI tool to print directory structure and file contents for LLM context generation

Static security scanner for LLM and agentic source code — finds prompt-injection, context-leak, PII, RAG, and agent-tool vulnerabilities before deployment.

A lightweight Python library providing easy access to popular LLM model names, metadata (context windows, capabilities, pricing tiers), and support definitions. Born from a failed experiment on context data compression for LLMs with finite context windows, this library helps developers manage model names, validate support, search by capabilities, and find models by requirements without hardcoding strings or maintaining spreadsheets.

Deterministic replay debugger for LLM agents

Fuzzy search LLM models from models.dev

The /llms.txt file, helping language models use your website

Lean MCP server for minimal-context docs via llms.txt

Low-code platform to build generative AI apps, chatbots and agents with your data

LLMstudio Tracker is the module of LLMstudio that allows monitoring and logging your LLM calls. By leveraging LLMstudio Tracker, users can gain insights on model performance and streamline development workflows with actionable analytics.

AST-aware llms.txt generator for Python and JavaScript/TypeScript codebases

Convert llms.txt to Claude Code Skill format

Python SDK, CLI, and Arena for 10 LLM providers. Compare models, synthesize a privacy-aware Best Answer, use MCP tools, fallback, caching, and direct BYOK access.

llmtest package

A new package that helps developers integration-test AI and LLM applications by validating structured outputs. It takes a user's test scenario or prompt as input, sends it to an LLM, and uses pattern

A package with useful functions for working with Large Language Models

Auto-track LLM cost, latency, and usage. Two lines of code, every provider.

An enterprise-grade LLM-based development framework, tools, and fine-tuned models

Client library for LLM Whisperer

Standalone CLI for llmwiki — entity-based knowledge wiki compiled from any directory of text files.

Thin proxy server for Claude Code and Laminar tracing

Generate and update agent config files from LM Studio models for VS Code Copilot, OpenCode, Pi, and Codex.

Real-time AI agent cost monitoring and budget alerting for local development

Maturity-aware harness manager for local AI coding agents (Claude Code, Codex CLI, Copilot CLI). Audit, init, and refresh AGENTS.md / GROUNDING.md / CI on GitHub Actions / sandbox at the right S0–S3 stage; ships an installable skill.

LAO — Pilot Mode plus planner→coder→verifier→reviewer orchestration over LM Studio / OpenAI-compatible APIs: structured tools, permissions, skills, hierarchical memory, optional MCP, SQLite state, and optional per-plan Git.

Local-first LLM wrapper (import local_llm_wrapper) with pluggable transports (Ollama, Apple, optional Claude Code CLI) and a stable text-in text-out API.

100% local RAG for Obsidian, Zotero, and Claude Code — LightRAG + Ollama + MCP

A generic, high-performance, low-dependency LLM programming framework inspired by dspy

A lightweight CLI task manager for AI coding agents

The AI Agent Codex — 15 production patterns, scaffolds for LangGraph/CrewAI/OpenAI Agents SDK, and a searchable knowledge base

Lowcode Agent Runner - 从导出的 JSON 配置加载并运行 Agent

Neural Retrieval-Augmented Generation for GitHub code blocks

Lumagent CLI - a terminal coding agent

agent framework designed to make AI agent development simple

Interface between LLMs and your data

macropragma - apply macros/substitutions/modifications on top of your source code

Magento Database Code end Media Dump & Table Analyzer Tool

One person. One AI company. Infinite leverage. Deploy FAANG-level AI agents named after Mahabharata characters to run your software company.

Claude Code automation agents for the MAID (Manifest-driven AI Development) methodology

Python SDK for mails-agent — email capabilities for AI agents

Maintainability gate and AI remediation prompt generator for CI.

Modular Python Agent Framework - Temporary Fork of mango for development purpose

Claude Code skill for Manim mathematical animations -- install with one command

Generating Mario Levels with GPT2. Code for the paper: 'MarioGPT: Open-Ended Text2Level Generation through Large Language Models', https://arxiv.org/abs/2302.05981

Autonomous development agent using AWS Bedrock and Strands SDK

mCodeGPT

mcodegpthippa

MCP Agent Graph - Python SDK for Agent Development Framework

Agentic code review MCP server powered by OpenAI-compatible models (defaults to free GitHub Models)

A specialized MCP server that tracks your Python development sessions, errors, fixes, and coding patterns using a persistent knowledge graph. This helps you build a searchable database of your development learnings, solutions, and insights.

MCP Server for AgentNEX - Bridge between NEXA AI and AgentNEX Backend

MCP implementation of Claude Code capabilities

Namespace alias for claude-command - AI command center for Claude Code

Namespace alias for claude-command - AI command center for Claude Code

MCP server for hot-swapping llama.cpp models in Claude Code sessions

Model Context Protocol server that proxies local Ollama to MCP clients like Windsurf and VS Code. Companion VS Code UI: https://github.com/pblagoje/mcp-ollama-extension

MCP server exposing OpenAI gpt-image-2 image generation and editing as a single tool for Claude Code and other MCP clients.

Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads

This package contains the code for training a memory-augmented GPT model on patient data. Please note that this is not the 'letta' company project with thehttps://github.com/letta-ai/letta; for use of their package, plsuse 'pymemgpt' instead.

Merlin — a fast local LLM for agentic coding on Apple Silicon

Codex app-server adapter and chatbot for Meshagent

A package for developing Meta Agents

Set of tools used for metagenomics research project

Set of tools used for metagenomics research project, using keras and pytorch

The Multi-Agent Framework

The Multi-Agent Framework

Production-ready AI agent framework with comprehensive testing, intelligent caching, connection pooling, unified interfaces, 36+ tools, and enterprise-grade security - 30-50% cost savings through smart optimization

Metis: local-first AI dev agent. Canonical types, event bus, adapters, routing, tools, memory, sessions, pricing, HTTP/WebSocket server, transparent provider-shape gateway, and CLI/TUI. Installs as `pip install metis-llm`; imported as `metis`.

Terminal-first coding agent CLI for local and OpenAI-compatible models

Nano SWE Agent - A simple AI software engineering agent

A language model fine-tuned for code and conversation.

A library for building LLM societies

A simplified AI Agent framework for learning Claude Code architecture

Minimal standalone coding agent scaffold.

Mini SWE Agent - A simple AI software engineering agent

A Minimalist Framework for LLM Workflows in 100 Lines of Code

Minimalist Library for programming with LLMs

A minimal PR review agent using LLM for automated code review

MiniMax-M2.5 AI terminal agent — chat, code, and create

Python SDK for building MIRASTACK agents — agentic workflow automation for platform engineering

Write your own code

Python Client SDK for the Mistral AI API.

Sparse AutoEncoder to decode Mistral LLM

Minimal CLI coding agent by Mistral

Minimal CLI coding agent by Mistral with WatsonX support

"It is a Machine Learning Copilot Agent that can execute code for end-to-end ML Cycle"

mlhook-agent - temporary placeholder

Ask your codebase questions using Ollama and Mnemosyne -- zero-config local code search

MCP server for Android device automation via ADB and uiautomator2

GPT development tools for Mojo

Monads implementation in pure Python code