Python agents

AI 辅助编程的安全与测试回归审查 Skill，分清主次守住权限/数据/核心业务风险 | Security & test-regression review skill for AI-assisted coding — triages auth, data-ownership & core-business risks

Local insurance-clause RAG agent: hybrid retrieve (BGE+BM25), cited answers, honest refuse. Not regulated advice. | 本地保险条款 RAG：混合检索、带引用回答、诚实拒答

A Script-Driven AI Game Master (Keeper) that manages long-context interactive storytelling for narrative role-playing games.

Evidence-first Agent Skill for economics and management paper writing, polishing, peer review, citation integrity, method safety, LaTeX, and auditable revision.

Local AI assistant for the Axelera Voyager SDK and Metis AIPU — semantic RAG over the public Axelera sources + workflow skills, via MCP. No account, no cloud.

面向企业用户的本地优先 AI 工作台：本地文件内容生成 + 工作自动留痕提炼 + 飞书文档治理，共享本地索引与 AI Agent，数据留本机、写回需确认。Local-first AI workspace for enterprise users.

Multi-agent control plane for cross-border e-commerce: LangChain + Claude ReAct workflows, RAG, and ERP/CRM-style tool integration. 跨境电商多 Agent 调度中枢：LangChain + Claude + RAG + ERP/CRM 工具集成。

Local-first AI context engine combining a structural Atlas, verified Archive, episodic Animus memory, CLI, MCP, and focused context routing.

Amazon 跨境电商运营助手 - AI 驱动的运营分析平台

Trading agent that learns from his mistakes

微信聊天记录 AI 本地桥 / WeChat History AI Bridge：基于 WeFlow，把本地微信 HTTP API 整理成可给 AI Agent 安全读取的 OpenAPI、脱敏 envelope、自检脚本和隐私边界。

让任何 AI 真正"看"视频：纯本地的视频阅读 skill——faster-whisper GPU 转写 + ffmpeg 场景感知抽帧，双通道带时间戳（t=MM:SS），数据不出机。1 小时视频 3~5 分钟读完。Give any AI agent eyes for video: 100% local video-reading skill — faster-whisper GPU transcription + scene-aware ffmpeg frame extraction, dual-channel with t=MM:SS citations. Nothing leaves your machine.

An open-source AI pipeline that transforms books and documents into production-ready Hermes Skills.

Voice teammate for Claude Code: talk to your live session hands-free, hear replies spoken. Local whisper STT + macOS TTS. /voice-on in any session.

Cost-aware multi-model orchestration for Codex CLI: route hard work to strong agents, replay proven paths with cheaper agents, and track cost, memory, and project evolution.

多 Agent 对抗式投研助手：多空各持一方举证、第三方裁决 | 自用工具，不声称盈利、不构成投资建议 | Multi-Agent stock analysis with Bull/Bear debate (Claude + DeepSeek)

强模型规划+弱模型执行+程序验证的编程任务调度系统。强模型只动嘴，弱模型干脏活，token 省约 60%。

AI Agent Skills library to facilitate technical lead in a software project

Claude Code skill for local LinkedIn lead scouting and outreach automation.

Kohaku-XinYu: 基于 KohakuTerrarium 的长期运行型个人 AI 伴随系统，支持记忆、学习、原生 QQ gateway 与受控主动性。

面向大学课程的证据驱动学习 Agent：连接预习、复习、错因修复、迁移验证与考前组卷，依据学习证据动态安排下一步。

Evidence-first workflow protocol for vehicle diagnostic AI agents.

基于 Dify + Ollama + MCP 实现的全本地化电商客服智能体

Agent-first OpenCLI skill with practical scenarios, desktop adapters, Chinese platform workflows, and troubleshooting guidance.

AI research workbench for observational cosmology that provably refuses to fabricate — every number traces to a registered dataset, or it abstains.

多模态 RAG 智能体：MCP 协议 + GraphRAG + Cross-Encoder 精排 + 多 Agent 编排 + RAGAS 评估

Custom AWS Transform agent that migrates PyTorch/Triton kernels to AWS Trainium NKI (@nki.jit) and compiles, numerically verifies, and profiles every candidate on a real Trainium device before opening a PR.

Self-improving SC2 Protoss bot: rule-based + PPO + Claude advisor, with autonomous code-level self-improvement loop.

AI 코딩 비서(Claude·Codex·Gemini)가 확인도 안 하고 '다 됐어요' 하는 걸 자동으로 막아주는 품질 안전장치 — 비개발자·바이브코더를 위한 한국어 우선 하네스.

Agentic Lemma Discovery for Program Verification in Rocq

Reversibility-aware safety signals for LLM tool-use agents | estimating action recoverability before irreversible operations: (delete/send/pay/commit) when rollback is unavailable.

Code for 'Evaluating Collective Behaviour of Hundreds of LLM Agents' (AAMAS SE 2026)

A scientific-method lab notebook your AI agent drives — research as a tree of questions and falsifiable hypotheses in plain markdown, with an LLM literature wiki and a browser GUI cockpit.

A runtime-native AI intelligence platform for portfolio analysis

A Python Telegram bot heavily run on Python that summarizes conversations within seconds using advanced LLM processing.

Dual-Gate-MoA: harness-enforced dual gates for LLM agents (recon before acting, QA before shipping) + MoA multi-candidate competition. Ships as a self-contained replication PRD.

Build self-evolving AI agent harnesses with portable harness units, artifact-aware testing, trace-backed diagnosis, and evidence-gated promotion.

A multi-agent system for sports match value analysis: an independent Elo power model vs. market odds detects mispriced outcomes (positive-EV value), with red/blue risk-control debate, backtesting, and multimodal input.

AI Data Analyst Agent , asisten cerdas berbasis Groq (Llama-3) & LangChain untuk eksplorasi, visualisasi otomatis, dan pembersihan dataset (CSV) langsung dari chat

Governed AI agents for education — 8 reference workflows on Amazon Bedrock with deny-by-default authorization, student-PII masking, human-in-the-loop enforcement, and tamper-evident audit trail aligned to FERPA, COPPA, IDEA/504, and ADA Title II. Built for K-12, community colleges, and universities.

This project focuses on developing a real-time, intelligent healthcare assistant powered by Agentic RAG (Retrieval-Augmented Generation) and the LangChain framework. Designed to provide multimodal support—text, image, and document-based inputs—it aims to assist medical personnel and individuals in remote areas.

Convert natural language into SQL queries using Large Language Models (LLMs), schema-aware prompting, and database validation.

RAG-Based Document Retrieval and Question Answering System

This AI system orchestrates multiple AI agents working together in a structured pipeline to generate comprehensive research reports on any topic.

Autonomous multi-agent AI system for intelligent GitHub PR review and bug detection. Built with LangGraph, RAG + tree-sitter, and GitHub MCP for deep codebase understanding. Automatically raises issues, drafts fixes, and opens PRs as smart-pr-review-bot[bot]. Supports Review Only, Human-in-the-Loop, and Auto-Pilot modes with full LangSmith tracing.

OpenGPT, An open AI workspace for conversations, knowledge, and productivity.

Seamless LLM Lineage for DataHub with LangChain and LangSmith

🧠 Build a RAG-driven chatbot to answer user questions using real-world knowledge from documents, focused on mental health and social policy in Indonesia.

Find where agent-controlled intent reaches consequential actions without an authority check. Python + TypeScript + Go. Zero-dependency SAST for AI agents.

Generate code and tests automatically using LangChain, Groq, and Gemini LLMs — all from a single CLI command.

Agentic CLI tool for investigating bugs using code and logs.

Run commands on network device with LLM using netmiko

LLM Sentinel Red Teaming Platform is an enterprise-grade framework for automated security testing of Large Language Models, detecting vulnerabilities such as jailbreaks, prompt injection, and system prompt leakage across multiple providers, with structured attack orchestration, risk scoring, and security reporting to harden models before production

This repository contains code examples of langchain

Trilith v0.1 beta — context engineering for agents (budgeted memory, 3 tiers, audit trail). Core is Python; use via REST or adapters (LangGraph / OpenAI Agents / Claude / TS). Needs trilith serve locally. No API key for Trilith itself. Not prod-hardened. Clone → pip install -e ".[server]" → try docs/adapters.md → open issues.

Autonomous Multi-Agent Environmental Command & Spatial Attribution Engine for 277 Municipal Corporations across India. Features real-time CAAQMS IoT streams, ESA Sentinel-5P satellite plume overlays, ML forecasting models, and automated compliance dispatch notices.

A governance framework for multi-agent AI systems. Structure, lifecycle, and quality controls for AI agent orchestration.

Neutral control plane for coding-agent fleets: claims, roles, mailbox reliability, receipts, audit, federation, dead-letter visibility, sandbox receipts, and cross-agent coordination. This is still pre-1.0.0 - Software works however you might need help from your local agents to set it up.

A deterministic, budget-aware control plane for Codex custom agents and Skills.

Automatic risk-based task routing and multi-agent orchestration for Codex using Sol, Terra, and Luna.

🤖 Enhance coding with Claude and Codex, a library for duo peer programming that combines AI strengths for superior code generation and reviews.

Your AI on rails, on your terms — an open-source Claude Code plugin that enforces every process as it happens, so your AI can't fake the work. Free, on the subscriptions you already pay for.

Competition-oriented brain + multi-worker penetration runtime built with LangGraph and OpenCode

AI-powered Board of Directors for multi-perspective strategic decisions. 10 director archetypes (Musk, Altman, Hassabis, Li, Nadella, Huang, Pichai, Hoffman, Ng, Jobs) + 7 shared methodology skills.

A股热门板块多智能体量化分析系统 | 基于FinGenius架构 | ETF50+60只个股扫描 | AkShare数据源免费

Evidence-based adaptive scaling and governance for homogeneous Codex agents.

Production-ready skills for Claude Code and multi-agent systems

Multi-agent HVAC operations co-pilot (dispatch, inventory, AR, risk). Phase 1 Streamlit demo with synthetic fallback + optional Mongo. Human-in-the-loop.

Jupyter notebooks, made agent-ready. 8+ formats · MCP server · async execution · token budgeting.

Autonomous content-publishing pipeline that learns from its own audience and rewrites its own prompts (agentic, self-improving)

Own Your Diff is a local human review gate for AI-assisted code changes. It generates grounded multiple choice questions before agents can push or open pull requests.

A full Google Cloud delivery team for your coding agent — vendor-objective solution designer, GCP + agentic architects, security, SRE, and QA — wired into a design-first, security-first delivery gate. Installable on Claude Code, Antigravity, Codex, and Kimi.

Fail-closed authorization gateway for agentic RAG — capability-label pushdown, deny-aware barriers, per-turn agent-context revalidation.

DataClawEval: A Benchmark for Engineering Data Agents in Real Industrial Harness

Codex-native operating layer for repeatable AI workflows with AGENTS autoload, project-isolated memory, plugin skills, and a thin ro CLI for terminal + VS Code chat.

The open standard for AI agent integrity. Evaluate, enforce, and prove that autonomous agents are adversarially coherent, environmentally portable, and verifiably assured.

Production-grade Agentic AI platform on Kubernetes — KEDA autoscaling, Kyverno governance, token budgets, OTel observability. Demo for CNCF Meetup talk "Agentic AI Without Chaos"

A science inference cluster — every model, from protein folding to LLMs, behind one OpenAI/Anthropic-compatible endpoint and a single key. Built to run beside HPC so agentic AI researchers can drive simulations and model inference together. Self-deploying on RKE2 with KServe/Knative + HAMi GPU.

A private, offline AI assistant running entirely on your local machine.

FastMCP 3.2 server plus Webapp for Yahboom Raspbot v2 robot car, ros2, raspberry pi

Companion code project for the ebook Vector Database Engineering for AI — Building High-Performance Retrieval Systems for LLMs, RAG, and Agentic AI.

Browse open-source tools for automating scientific research, from literature review and experiments to writing and peer review

A portfolio site, showcasing my blogs, repos, sites and applications. Built with Python backend and React/Vite frontend. Includes chatbot built using Gemini and Google ADK. Uses Gemini AI for content summarisation and metadata extract. Hosted on Google Cloud Run and Google Firestore.

Deterministic governance for Claude and MCP tool calls. Pin approved capabilities, detect drift, and refuse unsafe or unapproved actions before execution. No model in the decision path.

Agentic RAG system with LangGraph, hybrid BM25+FAISS retrieval, cross-encoder reranking, Corrective RAG, FastAPI, RAGAs evaluation, and Docker deployment

Agentic synthetic-data generation framework inspired by Meta FAIR's Autodata / Agentic Self-Instruct.

Docker-based, multi-tenant platform for autonomous AI agent ecosystems — runs on local or cloud models, with built-in cross-instance federation. Source-available under the PolyForm Small Business License.

Ultimate vibe coding plugins for Hermes AI agent — Effect-ts, LSP, Semble code search & Graphify knowledge graph + t/s status bar. 23 tools, 5 plugins, stdlib core.

Uses beads, souffle, z3 and mcps to give first-order logic powers to your thinking agents

A guard-railed, closed-loop workflow for AI coding agents: live state bus + execution-level hard intercepts for Claude Code and Codex (GitHub PR / GitLab MR). From step-level to requirement-level; eval-driven, spec-driven, human-in-the-loop.

Research on why autonomous AI agents don't know when to stop, and three engineered fixes.

A fully autonomous, agentic video generation tool for OpenWebUI powered by OpenRouter.

Free CLI tool for AI chatbot pentest. 6 attack classes (system prompt extraction, prompt injection, jailbreak). Single Python file. MIT licensed. From CVEasy AI.

Este projeto foi desenvolvido em contexto acadêmico por estudantes da Universidade Anhembi Morumbi – Unidade Mooca, dos cursos de Sistemas de Informação e Direito, com o objetivo de promover maior acessibilidade jurídica ao cidadão brasileiro.

Develop accessible ChatBot with Azure OpenAI and Streamlit

Simple Chatbot using LangChain just Q & A Generative AI Application.

This repo contains code for building a base telegram bot. Furthermore, on top of the base bot, you can add your own customization accordingly.

chatbot in python using chatterbot module and chatterbot-corpus for data training.

Corona bot using RASA

A simple whatsApp chatbot using Selenium library. Able to send messages to your contacts automatically.