Python agents

Claude Code plugins for reviewing changes: a four-pass code review, migration safety, regression hunting from git history, change-impact maps, tests from findings, and an orchestrator that runs what a change needs.

Multi-agent AI research pipeline: researches a problem area on the web, finds unsolved gaps, stress-tests them with a critic and majority vote, and designs solutions checked against existing work.

Manage your job search like a software project through GitHub with AI automation. Jobs are tracked as GitHub issues and projects. Agents scrape, triage, research open roles and update application status.

Evolution for the Agent Era. Inspired by Darwin, an open framework for experience-driven recursive self-improvement of AI agents.

Claude Code skill that delegates large reads and code writing to local models (Ollama) to save Claude tokens - with automatic checks, sandboxing and escalation. Never the cloud.

A curated list of environments, benchmarks, datasets, and RL training resources for workplace AI agents: email, chat, calendar, docs, and enterprise apps.

Spec-driven software-engineering team of three AI roles (Morfeo, Supervisor, Implementers) on Hermes Agent: Objective Contracts, isolated worktrees, independent review. v1.0.

AI Work Continuity & Cross-Model Agent Handoff System. Captures, verifies, and packages deterministic project state across Claude, GPT-4o, Gemini, Cursor, Antigravity & Local LLMs.

Pool multiple free-tier LLM API providers behind one OpenAI-compatible endpoint, with automatic failover and a live dashboard.

用自然语言查询经营数据的受控 Agent：模型只提议查什么，服务端核权限、执行只读 SQL、核实数字，回答里写明口径和依据。

Open-source AI cybersecurity companion

Gymnasium environment where AI agents solve SPaRC grid puzzles one move per step, with optional backtracking. Comes as a pip package, a Prime Intellect verifiers env and an OpenEnv server. The best model solves 16.0%, humans 98.0%.

FastAPI + LangGraph AI Agent backend with RAG, JWT auth, SSE streaming

No Code Multi Agent Orchestrator. Open source alternative to your favourites n8n, Gumloop, Relevance AI etc.

基于 RAG（Retrieval-Augmented Generation） 技术的服装领域知识管理系统。将服装搭配、洗涤养护、尺码推荐等知识文档向量化存储，支持通过 Web 界面上传新知识并自动去重。

OpenTelemetry-native agent observability: trace every LLM call, tool call and agent step, with cost per call, alerts, and Claude Code / Codex / Cursor session tracing. Python SDK.

FailSafe: An autonomous fact-checking framework leveraging Multi-Agent LLMs and Structured Argumentation Graphs (SAG) to verify claims with deep-web retrieval and reasoning.

context handover with increased security, brain child of RELAY_context

LLM plays a MUD and trains a world model based on the MUD

Reinforcement learning on LLMs in multi-step text-based environments.

A deterministic pipeline that diagnoses, explains, and safely auto-fixes failures in LLM agent systems.

Multi RAG File System using Groq, Huggingface, Llama Index, Langchain

AIcompanion è un assistente AI modulare sviluppato in Python, progettato per elaborazione conversazionale, analisi documentale e modalità di interrogazione. Integra Flask per l’interfaccia web, LangChain per la gestione del contesto e modelli Ollama eseguiti in locale fornendo un ambiente completo e ottimizzato per workflow AI on-premise.

A sophisticated, AI-powered trading system that combines traditional technical analysis with modern machine learning, multi-agent AI systems, and intelligent risk management - all wrapped in a beautiful, mobile-friendly dashboard! 🚀

The Customer Support AI Agent is a Python-based solution designed to monitor product prices on an e-commerce site. By utilizing an ARIMA model, it predicts future prices and sends notifications to customers when the price of a product is expected to drop in the next 10 days.

ReadmeCreator automates the generation of README.md files using graph-based workflow. Supports Ollama, OpenAI, HuggingFace and Anthropic models.

An Extremely simple agent framework just like OpenManus but much more easier to develop and faster to use

Local RepoGPT using Ollama + FAISS (RAG) + FastAPI Web UI. CPU-friendly AI assistant for querying code repositories.

Application for explaining difficult to understand textbooks based on Yandex Cloud API (YandexGPT-Pro, Yandex-OCR)

Memo MCP: intelligent journaling w/ LLMs, querying with RAG and data indexing. Retrieve past context for new conversations.

一个自动化密钥检测工具，对文件进行分析并提取潜在密钥。本项目获得DataCon2025软件供应链安全赛道第5名。

Multi-zone disaster relief AI env for Meta PyTorch OpenEnv Hackathon. 4-stage pipeline: PyTorch ZoneScorerNet -> Triage -> Planner -> Action Agent. False SOS detection, cascading failures, airlift precision.

Crucible Open Source (Basic): a minimal legal-agent orchestration framework (JP-law focused baseline), with mock-ready pipeline, prompt templates, and configurable adapters.

Cognitive architecture for emergent AI identity — blank slate to selfhood through lived experience

This project is designed to collect, analyze, and summarize news articles using AI. It integrates multiple news sources for collection and uses AI models such as DeepSeek and Gemini for analyzing the content. The system interacts with the AI to provide insights.

LLM weather agent built on the MCP protocol, combining Groq, LangChain and multi-API orchestration to answer natural-language queries with real-time data, forecasts and historical context.

SynapSense: Python In-Context Learning for Large Language Models SynapSense is a cutting-edge Python library designed to streamline the implementation of In-Context Learning (ICL) with Large Language Models (LLMs).

Scrapes newly posted jobs from a variety of sites and an ai agent filters them based off of the users resume.

A simple local python AI chatbot using HuggingFace inference API and streaming/history with Redis

AstrBot 宿主上的多 Agent 编排内核：TaskAnalyzer → AgentCoordinator → SubAgent 三级调度 + MCP Bridge + Shipyard 沙盒；435 单元测试 / 99% 覆盖 / ruff+mypy+bandit 全绿

A demo app showcasing different architectural patterns commonly used when creating chat apps or agentic workflows with an LLM

Various OpenAI projects and experiments implemented in Python. Utilizes all OpenAI APIs to test different models, fine-tuning approaches, and implementation patterns for production applications with comprehensive functionality tests.

RAG-Enabled Voice AI Agent for Calm, Clarity, and Everyday Emotional Guidance

DeterminAI is a deterministic, transactional AI execution engine that provides immutable, event-sourced workflows for AI systems.

OpenClaw Bootstrapping Benchmark is an automated test suite that evaluates how well local LLM models (served via Ollama or compatible OpenAI-API endpoints) can complete OpenClaw's first-run "bootstrapping ritual".

This is a File system mcp server that could allow an LLM to read and list files from a specified safe directory on your local machine.

Async Python implementation of the Claude Code agent harness — real Bash/Read/Glob/Grep tools, three-tier caching, multi-stage compaction, allow/ask/deny permissions, Pre/PostToolUse hooks, slash commands, subagent isolation. OpenAI + Anthropic + Ollama. ~1.9K LOC, 14 commits.

SAGE: Self-Adaptive Goal-directed Executor — A multi-tool LLM agent with DAG-based hierarchical planning, ReAct reasoning, and evidence-guided self-correction for automated research synthesis. Built from scratch, no frameworks.

A simple standardized way of providing and discovering downloadable skills that teach LLMs how to interact with a site.

Chat template support for training.

A high-performance, completely free and open-source command-line interface (CLI) that connects your local or cloud LLMs to the open web. It allows your AI models to search the web in real-time, pulling in actual search results to answer your prompts accurately without relying on outdated training data.

This project implements a travel chatbot powered by the RAG (Retrieve and Generate) chain, providing real-time information retrieval using various tools and the ability to fetch weather reports.

AskGPT is a real-time AI assistant that enhances your calls, interviews, meetings, and lectures by providing instant AI-generated insights. It captures live audio, transcribes speech into text, and generates context-aware responses—all in real-time!

Vision-grounded RAG agent for financial documents, built on a self-finetuned ColPali and LangGraph.

Research framework for adaptive, purposeful agents under uncertainty

Substrate examples

AI-Powered Appointment Scheduling Assistant (LangGraph v1+)

A single-file multi-client MCP gateway with per-tool access control.

An MCP server 🚀 for MonkDB, licensed under Apache 2.0 📜

A lightweight AI agent that answers natural language questions about APIs using their official documentation. It leverages vector embeddings and Retrieval-Augmented Generation (RAG) for accurate, context-aware responses.

📊 Transform research papers into professional slides and posters seamlessly and quickly with Paper2Slides, saving you valuable time.

A lightweight dashboard to view and analyze test automation results. Built with Streamlit + PostgreSQL, and powered by AI (Gemini) to help debug test failures faster.

Handcrafted ReAct Agent Engine | Supports FS / web page / weather query + code sandbox, including observability + visualization tools, focusing on circular thinking detection and task quality optimization

A project exploring agentic AI for Visual Question Answering. Compares zero-shot VLM performance (Transformers) against advanced pipelines using OpenCV (tool-use) and Chain-of-Thought (reasoning).

Cost-aware hybrid router for multi-agent LLM systems: keyword → embedding → LLM cascade. Matches LLM-only accuracy (82.6% vs 82.9%, McNemar p>0.3) with 74% fewer LLM calls on CLINC150.

Auditable benchmark framework for LLM trading agents with replayable trajectories, realistic execution, risk gates, and quickstart demos.

A modular framework for evaluating LLM agents in zero-knowledge text environments through semantic feedback and autonomous reasoning.

Do LLM Agents Dream of Better Hyperparameters? Comparing 5 frontier AI agents as autonomous ML researchers on Karpathy's autoresearch framework. 819 experiments, 7 hypotheses tested.

[SSBSE'25] This repository provides the artifact for the "KrakQL: LLM-Guided Blind Introspection of GraphQL Schemas" paper

Drop-in convention kit that enforces honest reporting in repos using LLM coding assistants — CI validators, smoke gates, PR templates, and schema-drift checks to stop scaffolds from shipping as features.

khojkar (pa: ਖੋਜਕਾਰ) is a deep research agent

LLM-driven autonomous power grid simulation and optimization using ExaGO. Large language models iteratively propose parameters, run simulations, and interpret results to solve complex grid analysis tasks.

Multi-agent due diligence pipeline that researches companies in parallel, producing a canonical JSON report with per-claim confidence, inferred source tier, and full cost/token observability. Markdown and PDF derive from the JSON. Built for provenance, auditability, and programmability over narrative polish.

The Execution Security Layer for the Agentic Era. Providing deterministic "Sudo" governance and audit logs for autonomous AI agents.

Markdown-first, local-first memory architecture for AI agents with lifecycle, distillation, and human-readable memory artifacts.

Persistent Anticipation Layer for LLM Agents — Pre-compute possible futures during idle time for faster, cheaper and more proactive decisions. Bio-inspired temporal awareness (short/medium/long-term horizons).

Provide tamper-proof, CMMC-compliant execution receipts that document AI agent actions in regulated environments for clear accountability.

Multi-Role AI Task Orchestrator — One task → 7 AI roles → One conclusion. Enterprise-ready with RBAC, Audit Log, Multi-Tenancy.

This repository contains a stateful AI shopping assistant for JD.com, implementing multimodal product search (text + image OCR) with LangGraph state management. Features Playwright automation, Gemini/Qwen-VL integration, and Gradio UI.

Agentic Design Patterns for LLM Workflows — A curated collection of reusable, modular patterns for building robust and autonomous LLM-powered agents, covering multi-agent orchestration, memory, prompt flows, and tool integrations

Deterministic, self-improving AI agent execution harness and runtime for the AXIOM ecosystem. Enforces plan-deterministic execution, parallel swarm orchestration, and signal-level sandboxing.

pm-agent (Octopus Assistant / 章鱼助手) — a multi-agent dev orchestrator that, like an octopus, dispatches tasks across multiple coding CLIs (Claude Code / Codex / Gemini) in parallel.

Create AI agents from urls or a documents csv file.

Developmental lifecycle architecture for AI agents: childhood stages, mortality, sealed wisdom distillation, soul persistence, and relational witnessing. From mountain night shifts to open garden.

Standardized Test Case Generation for LLM Agents via Model Context Protocol. Bridging the gap between AI and QA (Xray/Jira).

Qualitative benchmark suite for evaluating AI coding agents and orchestration paradigms on realistic, complex development tasks

A framework for supervising reinforcement learning agents using large language models.

AI-driven trading agents for financial markets

An automated AI research-code writer and AI research-paper writer implementation through a skills+ autoraters using any coding agent (Claude Code, Cursor, Antigravity, Cline, Aider). No API keys, no LLM SDKs.

Research constitution and specification framework for comparing independently governed AI judgments without transferring authority.

A growing collection of AI projects exploring LLM, Agents, RAG using LangChain, Gemini, open-source models, and more. Dive into cutting-edge use cases and collaborate to create the next big AI innovation!

Just A Rather Very Intelligent System.

Kazi Core is a self-hostable, open-source agentic engine. Plug in your own tools, run on your own infrastructure, and keep full control of your data.

Lightweight Personal AI Assistant for Git & GitHub in your CLI

A running multi-task multi-data terminal science benchmark for LLMs in GW astronomy

Toolkits for pydantic-ai agents: FS, SQL, pandas, memory, RAG.

AI-powered financial news analysis capabilities.

This project is an ATS Resume Agent that optimizes resumes for Applicant Tracking Systems. It uses Flask + CrewAI backend to clean, rewrite, and evaluate resumes against job descriptions, and a Next.js frontend to present polished resumes, ATS scores, missing keywords, and actionable recommendations in a user-friendly interface.