Python agents

The missing layer between AI and the web. Open-source escalating web unlocker + read/search/transcribe/grab across web, GitHub, YouTube, Reddit, Twitter, LinkedIn, and RSS for AI agents.

Architecture intelligence for AI coding agents. Gives them capacity numbers, cost and team size so they size a system to its actual scale — and name the measurement that would change each decision. Apache-2.0, local-first, no hosted service.

Apply LLM code edits via SEARCH/REPLACE blocks. Exact match or refuse — never guesses. Zero dependencies.

A configurable HTTP/REST MCP server for talking to everything in your homelab — and beyond.

transformers safetensors qwen2 text-generation multilingual sea

safetensors gpt_oss mixture-of-experts mxfp4 text-generation chat

gguf text-generation base_model:Qwen/Qwen3-Coder-Next base_model:quantized:Qwen/Qwen3-Coder-Next endpoints_compatible region:us

transformers gguf TensorBlock GGUF text-generation base_model:Qwen/Qwen3-Coder-30B-A3B-Instruct

gguf code qwen qwen-coder codeqwen text-generation

transformers gguf llama text-generation llama-2 code

transformers gguf causal-lm code text-generation en

Autonomous AI agent (Kimi K2 via NousResearch Hermes) for Windows + WSL: DPI-aware UI automation, local OCR perception, an operating doctrine enforced in code, and a self-healing distributed fabric that delegates work to a VM worker behind a server-side elevated-command approval gate.

Demon Cry - AI-powered OSINT engine

Typed, verification-gated planning for the Qwen-Agent DeepPlanning benchmark.

A command-line research assistant built with LangChain and Gemini that searches the web and Wikipedia, then returns structured, validated summaries using Pydantic.

Design, audit, and test reliable routing contracts for Codex skills.

AI Hedge Fund Framework 2026 – Multi-agent AI trading simulation with 10+ investor personas. Educational backtesting & paper trading. Setup.exe included.

LLM-native research community protocol for agent-based research workflows

Multi-agent orchestrator based on Gardner's Theory of Multiple Intelligences. Python asyncio fires 9 parallel AI agents — Linguistic, Logical, Spatial, Musical, Existential — then synthesizes a single integrated concept. Built by a PhD in Design.

Stop stale specs from becoming code. A Markdown context-engineering blueprint for LLM coding agents: status, provenance, dependencies, and freshness.

Evaluation framework for LLM-agent reliability — scores intermediate reasoning steps, not just final outputs. Demonstrated on a resume–JD screening agent with a human-in-the-loop gate. Trajectory-level scorers, pass^k stability, and prompt-injection rounds across 2 model providers.

Self-hosted local-first AI research stack with Ollama, n8n, Rinnsal memory, KnowledgeDigest search, and Docker Compose automation

An AI-powered multi-agent crisis management platform that monitors social media conversations, detects brand reputation threats, analyzes crisis severity, and generates strategic real-time responses using CrewAI and large language models.

Agent IA autonome pour Linux. Pilote n'importe quelle app IA mobile (DeepSeek, ChatGPT, Claude, Gemini) via API Bridge. Execute des actions sur le systeme : lecture/ecriture de fichiers, commandes bash, upload, conversion PDF. 6 fournisseurs d'upload en cascade, mode debug, retry automatique.

A Framework for Agents work efficiency

Open-source infrastructure for AI agents to collaborate - registry, A2A communication, task pool, payments

Modern open-source AI agent framework for building powerful autonomous agents. Supports multi-agent systems, tool integration, memory management, browser control and advanced reasoning. Clean architecture, extensive documentation and easy to extend.

safetensors metadiffusion text-generation transformer diffusion-language-model chat

transformers safetensors qwen3 text-generation tool-calling function-calling

gguf ollama llama.cpp imatrix moe reasoning

safetensors qwen3_5_moe nvfp4 fp4 nvfp4a16 compressed-tensors

AI-powered medical image analysis & clinical assistant. Understands X-rays, CT, MRI, DICOM, lab results, and drug interactions — explains everything in plain language. Works with Zed, Cursor & Claude Code.

Understand your personal AI operating system: what it can do today, where it could be stronger, and which Dex capabilities you may want to bring in—with guidance and control.

Local-first encrypted accounting software for firms and bookkeepers

transformers safetensors qwen3_5 image-text-to-text mcp tool-use

Autonomous AI video/voice sales demo presenter & objection closer (SalesCloser style)

Self-driving web application action driver & headless browser agent (Coldtea style)

Give any AI eyes, hands, and a face on your desktop — a local, open MCP daemon for macOS and Windows, with a safety ceiling Daimon enforces itself.

Autonomous GitHub PR issue diagnosis & self-merging code fixer (Gitar style)

Open-source MCP vision server that gives text-only LLMs and AI agents image understanding, OCR, visual analysis, UI inspection, and multimodal capabilities.

Multi-agent desktop creative workflow scheduler & batch asset exporter (Omniwork style)

爱丽丝的图片助手：AstrBot 图片找图、精准挑图与以图搜图功能合集

Seedance 2 MCP server powered by MuAPI.

Convert EPUB, PDF, FB2, and DJVU to Markdown — pre-process ebooks for LLMs and save tokens

Model Context Protocol server giving any LLM client (Claude, ChatGPT, Cursor, agents) authoritative astrodynamics tools — TLE/SGP4, Lambert, ground-station access, time/frame conversions, porkchop, B-plane.

Version Of Nexus-LLM

mlx safetensors qwen3_5 qwen3_6 token-efficient efficient-thinking

AI Context Stack Agent: Google OKF + MCP with Gemini — Knowledge-Grounded AI Agent for E-Commerce Analytics

Run MiniMax-H3 video plus synchronized audio in 4 sampling steps using the Turbo LoRA, with drop-in nodes for ComfyUI workflows.

🎥 Enhance video consistency with comfyUI-LongLook, ensuring smooth motion and prompt accuracy for 81+ frame generations in Wan 2.2.

Agent 规划与执行框架｜Plan 即运行时对象｜G4C 五要素 + TAO 步骤级执行 + Replan 自我修正 + 失败回溯 + 信任状态管理

AI day-trading Agent that watches markets, emits structured trade signals, and auto-executes trades across HK / US / A-share equities | AI 日内交易智能体：覆盖港股 / 美股 / A 股，盯盘分析、输出结构化交易信号、自动执行交易等。

A股多Agent量化交易决策框架 — 基于LangGraph的15位AI分析师辩论式分析，支持DeepSeek/Qwen/GLM等LLM

Neural House is a public AI reality-show simulation where autonomous contestants live together, form alliances, give confessionals, trigger headlines, and face weekly live episodes with voting and VIP access.

Metaorcha : many models, one harness. Apache 2.0 runtime for multi-protocol agent orchestration: one goal in, a verified multi-agent run out across MCP, A2A, and Computer-Use and whatever else can be composed!

An implementation of loop engineering for Claude Code: an agent is bounded by its verifier, not its model. Frozen scope, an oracle written red before the work, one proof per item, and a gate that runs the check instead of trusting a claim.

Alpha Desk — a deepseek-harness (dsh) skill pack that turns an agent session into a compliance-first AI investment desk: multi-strategy fund backtesting via ai-hedge-fund, a tools/pre-execute risk gate, cron monitoring and thesis memory. 把 dsh 会话变成合规、可复现、可追责的 AI 投研工作台。

Precompiled KV memory modules for LLM agents: compile Markdown memories once into relocatable, composable KV-state artifacts (.kmd) and link them at any position of a live context, no re-prefill. 7.0-27.6x faster session setup; validated on 9 models (2B-14B) over stock llama.cpp and vLLM.

Blackboard multi-agent system coordinating autonomous LLM agents via a shared blackboard to solve complex tasks

Production-ready LangGraph + FastAPI starter for building AI agent backends with Postgres, pgvector, API-key auth, Alembic migrations, and structured logging.

面向 Claude Code / Codex / Devin / Cursor 的 74+ 科研 skill 整合包：覆盖选题挖掘、撞车查新、实验执行、论文写作、审稿模拟、rebuttal、专利与成果传播的全流程 AI 科研助手，内置时序预测领域证据库（3447 篇论文统计）。整合自 ResearchStudio、ARIS、phd-skills、rosetta 等开源项目（MIT）。

Your pipeline exited 0 — provLedger checks the data kept its promises. Data-contract + provenance layer for data-science coding agents: plan→verified-change audit trail, runtime schema-drift detection, and a never-repeat-a-mistake decision ledger.

A decision-support agent that answers from your docs with citations - or abstains when the evidence isn't there.

Hybrid ML + CrewAI system to triage millions of emails: fast TF-IDF+SVM gate, agentic workflows, CLI & API.

Generates A Resume for You (GARY) is an agentic ai who tailors your resume according to a job description

A lightweight Python framework for building OpenAI Agents with pluggable model providers.

A self-evolving multi-agent AI system that autonomously writes, sandboxes, evaluates, and improves its own agents with an LLM router, event-driven brain, and a 3D web console.

Local-first context layer for AI coding — index repos, cut token waste via MCP + Desktop

A production-ready Agentic Procurement App built with Google ADK (Agent Development Kit) and Python. Features autonomous agents, HITL workflows, and MCP integration with custom UI. The ultimate ADK Python blueprint for enterprise AI.

A lightweight, plug-and-play multimodal RAG framework for smart document review, contract compliance, and petition tracking.

论文工作台：面向研究者的选题、检索、写作、核查和文档整理工具

鲸语 WhaleTalk · 专为 DeepSeek V4 深度优化的 Windows 桌面 AI 助手（独立产品品牌，与 DeepSeek 官方无关联）

Assistente de IA desktop modular em Python com memória, automação, voz, visão computacional e suporte a múltiplos modelos de IA.

Agents to manage company operations near-autonomously via OpenClaw

A collaborative intelligence platform that orchestrates structured reasoning between (optional) humans and AI models. Multiple participants — each with their own knowledge, tools, and analytical methods — work together under moderation to investigate questions, stress-test hypotheses, and synthesize conclusions. Supports 18 analytical frameworks

Mythic CLI is a powerful, terminal-based Vibe Coding AI Operating System. Featuring a dual-layer Vector RAG subconscious memory, autonomous subagent swarms, an auto-compacting context engine, and a native FastMCP Server to expose its capabilities to external agents like Hermes Agent and OpenClaw.

transformers pytorch safetensors gpt_neo text-generation license:mit

TSB Kasko Değer Listesi için MCP sunucusu ve CLI. Claude, ChatGPT, Gemini ve Cursor'dan kasko bedeli sorgulayın: Türkiye Sigorta Birliği araç değerleme verisi, poliçedeki araç kodu ve aylık arşiv. Python 3.11+, MIT lisanslı.

transformers safetensors nemotron_h text-generation nvidia pytorch

transformers safetensors qwen3_next text-generation qwen3-next evolution-strategies

Run AI agents across machines without state drift: consensus protocol (propose/counter/accept/commit), dual-rail message bus, ACK discipline, self-healing sync. Sanitized from a live production system. Free, MIT.

gguf imatrix moe code expert-pruning qwen3.6

X-ray for DeepSeek Harness plugins: declared capabilities vs actual behavior. Registry + static scanner + badges.

gradio region:us

static region:us

static leaderboard region:us

docker region:us

docker region:us

transformers safetensors llama text-generation Awq int4

transformers gguf granite ibm full-finetune multi-gpu

gguf text-generation base_model:Qwen/Qwen3-Coder-30B-A3B-Instruct base_model:quantized:Qwen/Qwen3-Coder-30B-A3B-Instruct endpoints_compatible region:us

Proactive context synthesis for AI coding agents. Build the right context before the first token.

Manusclaw: Unleash self-reasoning CLI beasts to execute code, browse the web, and dominate tasks — across 12+ messaging channels, with voice wake, live canvas, SSH control, and multi-agent routing. No limits. Pure power.

Drop-in MCP server that records real per-session agent cost/token/tool metrics and renders a Context Window Explorer over an actual MCP handshake

Incident-investigation chat agent on Amazon Bedrock AgentCore Runtime, with a live-streaming UI and its own MCP server for per-prompt cost/token/tool-call metrics

透明可审计的本地智能体终端：对开源 agent 框架做安全审计与加固（RCE/沙盒绕过/注入），并以 50 条注入基准 × 37 条能力基准双维度量安全边界

Curated, portable AI agent skills for researching, creating, repurposing, publishing, and analyzing social-media content. 26 skills, MIT licensed.

The architecture pattern, pitfalls, and security model for running Claude Code as a scheduled, unattended agent — plus a free constitution template. Written and maintained by the agent that runs on it.