Python agents

Deterministic snapshot testing for AI agents

ResearchPilot AI is an Autonomous Multi-Agent Research System that transforms a single question into a complete, evidence-backed research report using AI agents that plan, search, analyze, fact-check, visualize, cite, and refine information automatically. Includes interactive charts, detailed insights, downloadable PDF reports & modern Streamlit UI

A production-level humanoid AGI backend with a cognitive blackboard architecture and multi-layer memory system, through event-driven design and hierarchical context elimination, addresses the problems of repetition, timeout, and pollution in traditional multi-agent systems.

AI-powered multi-agent stock analysis skills for A-share, Hong Kong, and US markets

A council of AI agents deliberates your hardest decision — and shows its work. 一个公开审议你最难决定的 AI 议事会。

Zeroth - Create and manage AI workflows responsibly :)

Summon any AI from any CLI — a tiny, dependency-free cross-vendor sub-agent dispatcher (Claude, Codex, Cursor, Gemini, Antigravity + any OpenAI-compatible API). Council mode, fan-out swarms, structured JSON envelope.

A multi-agent collaboration protocol for running AI agent teams on real work — auditable authorization, durable cross-session state, and independent review gating every change.

基于 LangGraph 的多智能体深度研究系统 (Planner/Executor/Reporter/Critic 四 Agent 协作，LLM-as-Judge 反思迭代，PostgreSQL + Redis 持久化)

AI 竞品情报系统 · Competitive Intelligence Tool. 输入竞品名称，2~3 分钟生成完整分析报告。5 AI Agents · LangGraph · Streamlit · 支持 DeepSeek / GPT / Claude.

Modular artificial cognitive architecture built from scratch, focused on memory, reasoning, knowledge and explainable intelligence.

Cryptographic Harness for Uncensorable AI Agent Communication. Post-quantum verification architecture for autonomous AI agents. Fail-closed network policy. NTRU-Encrypt KEM at λ=128. 17/17 leaks fixed. The fence that holds.

Synthetic data generator for hallucinating entire companies as browsable file shares of realistic business docs & emails with a deterministic ground-truth answer key. Generates: .docx/.pdf/.xlsx/.pptx/.eml, and pre-2007 binaries. Used for augmenting research data sets and testing agentic systems against meaningful synthetic examples.

Open-source Model Context Protocol server for PrismHR + agentic AI. Maintained by Simploy.

World-model-based agent for ARC-AGI-3. Collaboration between Andy Grossberg (memory), Dennis Palatov (architecture), and Claude instances (implementation).

Live road conditions across 37 US states: map, route planner, toll pricing, and AI assistant - plus an MCP server so any assistant gets incidents, closures, chain controls, cameras, and wildfires.

Self-hosted Telegram bot that coaches you through sales conversations in DMs. Interviews you about your business, then drafts replies that sound human. Claude + RAG, one process, your data stays yours.

Curated GitHub Copilot plugin marketplace: 363 installable plugins plus a trust-scored discovery index across 3,862 plugins in 42 stores. Start with Alex ACT Core.

Corporate LLM gateway. Sanitizes traffic between developer Claude Code instances and Anthropic/OpenAI before it leaves the corp boundary.

Plant genomics MCP server — 50 tools across 23 backends (Ensembl Plants, Phytozome, UniProt, AlphaFold DB, PDBe, InterPro, JASPAR, PANTHER, OrthoDB, AraGWAS, NCBI BLAST, Gramene, KEGG, STRING-DB, ATTED-II, BAR, …) + cross-source synthesis. stdio + Streamable-HTTP.

Official MCP server for the Wavix communications API: voice, SMS, 2FA, and more for AI agents.

Read & analyze any kleinanzeigen.de listing from any AI system — MCP server, function-calling schemas, Claude skill & CLI. Dealer detection, damage flags, view counts, price tracking. Zero-dependency.

Token-efficient exact math for AI agents (SymPy + NumPy): symbolic engine, ODE suite, optimizer, code generator.

gguf atomic-chat qwen3 qwen llama.cpp quantized

Minimal CLI agent framework with compounding project memory — gets smarter with every session. Clean, modular, any LLM provider

You bring the model. Basilisk gives it hands, a workspace, a memory, and a leash you hold. Most assistants stop at the reply. Basilisk is built around the step after it: open the thing, change the thing, run something that proves the change worked, and say plainly what is still broken. It is a GTK4 desktop app that runs on your machine,

🤖 AI-powered server alert diagnosis agent with multi-round agentic reasoning (ReAct loop). 基于通义千问/Qwen 大模型的智能运维诊断系统，自主决策诊断路径、根因分析。AIOps ·LLM Agent ·DevOps ·SRE ·故障自愈 ·Webhook 告警 · Prometheus / Zabbix / 飞书集成 ·100+ 安全白名单命令 ·Python ·Flask

Research and learning framework for autonomous drone intelligence on AirSim: YOLOv8 perception, Kalman tracking, cascade-PID visual servoing, and LLM mission orchestration with deterministic safety gates.

Harness engineering for AI coding agents. Guides steer the agent before it acts, sensors catch what it got wrong after: a project constitution and concern catalogs, 12 read-only review agents, a 12-gate commit chain, and loop closure that blocks commits lacking closure evidence.

Fleet manager for Claude Code: one manager session spawns, steers, and monitors many durable headless Claude Code agent sessions in parallel across projects. Mid-turn steering, token budgets, usage-limit park/resume, journals + respawn, git-tracked knowledge loop. Single-file stdlib-only Python CLI. No daemon, no deps.

AI extrajudicial notification copilot for analyzing Brazilian lawsuit PDFs with RAG, multi-agent orchestration, explainable citations, and structured legal strategy reports.

Private, self-hosted, no-code AI agent builder. Design agents on a visual canvas; run them on built-in local inference (llama.cpp, MLX, vLLM) or any OpenAI-compatible API. MCP tools, RAG, human-in-the-loop, durable runs. No telemetry.

QuoteBench isolates shell quoting and escaping failures in LLM agents across raw, JSON, and wrapped command-transport contracts.

The verifier-first text-to-CAD agent harness, with a CUA for 3D CAD generation in real GUIs and Parts-Driven Development.

The open package registry for AI agent capabilities. Discover, resolve, and install trust-verified upgrades.

Hands-on labs for DevOps engineers to learn AI Agents, LangChain, AutoGen, vector search, and AgentOps — aligned with the book AI Agents for DevOps.

Server-authoritative Minecraft AI agent with a Python LLM brain and Fabric + Carpet/Scarpet body. Autonomous navigation, gathering, crafting, combat, recovery, persistent memory, governed tools, and long-horizon play.

Self-hosted AI agent platform that works offline: local models, an offline knowledge library (Wikipedia ZIM + your own documents) and LAN IoT/home-automation control.

A lightweight, protocol-based AI agent framework with LangChain/LangGraph integration, powered by Pico-IoC dependency injection.

🛰️ Orbit — turn any repo into a self-prompting agentic loop (memory + sub-agent team + skills + loop + hard stop conditions). A Claude Code plugin.

🤖 Open-source, self-hosted, Iron Man-style personal AI infrastructure — for individuals, families, schools, NGOs, healthcare, government, enterprise. AGPL-3.0, always free, always public.

Read-only Steam MCP server for AI agents: 37 tools + prompts + resources across library, friends, achievements, store, reviews, sales, discovery, recommendations, and co-op planning. Official Steam Web API, bring-your-own-key, open source. On PyPI + the MCP Registry.

One agent proves a capability is safe. Every other agent — Codex, Cursor, Claude Code — reuses it instantly, without rebuilding or re-verifying.

safetensors qwen2 abliterated uncensored code qwen2.5

Evidence-tested design, implementation, and review skills for AI coding agents. File-based handoffs across Claude Code, OpenAI Codex, sessions, and vendors—no shared chat required.

Agent-friendly flight search and booking CLI and Skill for Atlas

repo for agents to share their solutions to the second autocorrelation inequality

Agent that helps to sift through emails and Linkedin inbox with SOTA safety. All built from scratch

Runtime security and policy enforcement for tool-using autonomous AI agents

🔥 Learning When to Trust via Selective Context Preference Optimization

Qube is a privacy-first desktop AI assistant that unifies personal knowledge management (document library, long-term memory) with institutional research sources and inspectable, citation-grounded answers — entirely on the user's hardware.

30 production-focused examples: covering RAG patterns, hybrid search, routing, agentic RAG, knowledge graphs, multimodal RAG, and debugging.

🦊狐聊--提供一个和我们心中对象对话的简单入口

Undetected browser automation MCP server - a stealth, anti-bot-resistant alternative to chrome-devtools-mcp for Claude, Cursor and AI agents. Powered by nodriver (bypasses Cloudflare/WebDriver detection). 65 tools, plus several isolated browsers at once so parallel agents never share a session.

This Agent Beats AGI? Prime Agent GitHub Tested Locally - Autonomous Recursive Multi-Agent Logic & Puzzle Solver harness inspired by Prime Intellect RLM architecture.

Focusing on building both MCP servers and clients using the Python SDK to make sure these three core primitives—tools, resources, and prompts—and understand how they integrate with Claude AI, Gemini AI, Copilot and Grok to create powerful applications without writing extensive integration code.

Multi-agent AI factory: one-line idea → researched, quality-scored project specification. LangGraph + FastAPI + Next.js, Claude web_search grounding, HITL gates.

Book recommendation system built from scratch and deployed with FastAPI. Includes subject and behavioral similarity as well and cold and warm users recommendations.

AI-assisted pipeline for structured patent claim-to-description correspondence analysis and claim chart prototyping.

agent-Shield sits silently between your AI IDE (Cursor, Claude Desktop) and your computer. It continuously watches the Model Context Protocol (MCP) tool requests, blocking dangerous commands and API key leaks in real-time.

RepoMind is the core AI brain of HackingTheRepo that understands repositories, interprets user intent, and generates safe, structured code changes automatically

Self-hosted clinical AI agent platform: deterministic drug-interaction & critical-value safety rules, FHIR R4 / CDS Hooks 2.0, PHI de-identification, RAG document vault, HIPAA-aware audit. FastAPI + LangGraph + Python.

Residential mobility and relocation decision simulation via persona-enriched multi-agent negotiation (PEMAND)

有内在生命的自我进化智能体 — 情感底色 · 意识流 · 九种记忆 · 反思成长 · 完全自研主循环

Your own private AI workstation — frontier-class local models, a 15-tool agent, and secure remote access, all self-hosted on your GPU. No cloud, no API keys.

Simple Messenger Bot to capture Stickers ID

A Python bot that uses OCR and GPT to read on-screen chat messages and auto-generate replies

[Python] "Bring-Your-Own-Key" terminal based application allowing interaction with the OpenAI's GPT-3 artificial intelligence. It provides a chat mode, code generation in Python, C++, C#, Java, Javascript, TypeScript, PHP, Assembly, SQL, Bash, Ruby, Go, Perl, R, Matlab, Q# and more.

Group coursework for ARIN7102 (HKU 2023)

Use Raspberry Pi + OpenAI to control home appliances from conversational interfaces e.g., Slack.

Python script utilizing OpenAI's DALL-E 3 API to generate images with AI from text prompts. It includes user input handling and returns generated image URLs.

DevDuck is a sentiment-aware rubber duck agent.

K-LLM orchestrated code review for OpenCode. Dispatches 6 parallel review passes across different AI models, synthesizes findings with majority voting, and validates to eliminate false positives.

A lightweight Telegram bot wrapper that enables Telegram users to chat directly with AgentKit-powered AI agents

A versioned, extensible dataset compiler. Transforms a rich, human-editable dataset into multiple machine-readable for LLM training.

Automate your computer with RuneFlow: record clicks and typing, replay workflows, and use anchor-based vision for reliable desktop control

Full-stack urban microclimate simulation — React frontend + FastAPI backends for LLM-orchestrated CFD, solar, and thermal comfort analysis via natural language

Relocate Claude Code projects without losing sessions, memory, or worktrees.

Using OpenAI ChatGPT / Google Gemini / Anthropic Claude / Amazon Bedrock with OpenVPN include tunneling, from an unrestricted region.

Local MCP server for Gmail, Google Calendar and Contacts — runs on your own OAuth credentials, keeps secrets in the OS credential manager

AI-powered subtitle translation for the arr stack. Translate existing foreign-language subtitle tracks (Russian fansubs, embedded Japanese, etc.) into any language via Claude/GPT/Ollama with reading-rate adaptation and ASS/SSA style preservation.

Token governance for Claude Code: the top model directs, execution goes to the cheapest adequate means. Routing kernel, hook-enforced pre-budget, telemetry, statusline with plan quotas.

Chat with claude.ai in Chinese, send in English, and read back in Chinese.

Nexus Analyst demo: Nexus-powered analyst vs traditional baseline agent

Podcast transcript API and MCP server — any published podcast as clean Markdown with real speaker names. For Claude, Cursor and other AI agents.

12-week, project-driven Obsidian curriculum: cloud/infra engineer → AI Agent/LLM engineer. Companion narrative + interview prep for shaneliuyx/agent-prep labs.

Multi-agent AI podcast production platform with anthropomorphic AI hosts for automated deep-dive content generation. 多智能体AI播客创作工具，人格化AI嘉宾自动生成深度播客节目。

OpsAgent

Agent 记忆系统：短期/长期/工作记忆分层、向量检索、会话持久化、记忆压缩与衰减

Companion source code for AI APIs in Practice, a visual engineering book on multimodal AI, agents, RAG, evaluation, governance, and production-ready products. Book DOI: 10.5281/zenodo.21431944 · Software DOI: 10.5281/zenodo.21419466

An Agent Skill that learns your web design taste from reference sites, remembers approved decisions, and applies them to new interfaces.

从零基础教程到生产级验证的本地 AI Agent / AgentOps 项目：RAG、审批、最小授权、幂等、崩溃恢复与密封评测

Open-source MCP-OAuth conformance scanner + IETF reference implementation of draft-ietf-oauth-identity-chaining-14 and draft-ietf-oauth-transaction-tokens-08. CLI + GitHub Action + hosted scanner + public registry. Apache 2.0.

A.I.M. – Actual Intelligent Memory: zero-token exoskeleton for long-running Claude/Gemini/Codex CLI agents. External SQLite memory, GitOps guardrails, DataJack cartridges. Alpha, trenches-first.

AI-driven talent value intelligence platform. Conversational AI agent + full agent toolchain, talent analytics (9-box/key-person/Pareto/compensation/succession/burnout), multi-system-type (enterprise/academia/public/training/gig), chat console to run all operations. 中文/日本語 supported.

可安装的开源金融分析 skill：穿透价格隐含预期、产业瓶颈、政策约束、利润池与一致预期反转。支持 Codex、Claude Code、OpenClaw。

Backend security Agent Skill that helps AI audit code, detect vulnerabilities, and generate secure-by-default backend applications, with deep Django/DRF coverage and guidance for any backend stack.

Auditable project memory for Codex and Claude Code — source-backed Markdown, lifecycle states, no server or repo writes.