Python agents

Berkeley Function Calling Leaderboard (BFCL) with Chinese-Language Evaluation

AI agent for creating personalized digests of research papers

LightweightEmbeddings is a fast, free, and unlimited API service for multilingual embeddings and reranking, with support for both text and images and guaranteed uptime.

StickerSelector 是一个基于语义向量匹配的表情包选择系统，用于让 AI 在聊天中发送真正符合语境的表情包。 与传统依赖关键词或规则的方案不同，StickerSelector 会将 AI 生成的“表情包意图描述”与本地表情包进行语义匹配，从而选出在当前语境下最自然、最像真人会使用的表情。 适用于 AI 聊天机器人、QQ或微信的自动聊天、AI 女友等需要“拟人化表达”的场景。

📚 AI-Powered Book EPUB Knowledge Extractor & Summarizer Transform your PDF books into structured knowledge effortlessly! This tool leverages AI to analyze books page by page, extracting key insights, definitions, and concepts, and organizes them into Markdown summaries for easier study

Real-time behavioral intelligence for call centers. Transcribes support calls, redacts PII, extracts emotional tone, classifies issues, and delivers insight-rich dashboards — powered by GPT-3.5 (cheap tokens), Whisper, DuckDB, and a polished React+TypeScript frontend. No Azure. No Power BI. No vendor lock-in. Just full-stack AI that runs local.

🕰️ Time machine for Claude Code sessions - Automatic snapshots, granular rollbacks, and visual diffs. Never fear AI mistakes again.

Zero-config MCP server for searchable documentation (SQLite default, PostgreSQL optional)

An open-source, self-hosted API that turns standard email providers (Mailgun, SES, SendGrid) into "Inbox-as-a-Service" for AI Agents.

Knowledge-base-driven article generation, fact audit, and 7-platform publish packs for WeChat, Xiaohongshu, Zhihu, Juejin, CSDN, Toutiao, and Weibo, with Wechatsync draft-sync handoff.

Infrastructure Context for your Coding Agents (Claude Code / Cursor)

Curated technical knowledge base: 1287+ dense reference articles across 28 domains. Agent-first format with wiki-links, gotchas, runnable code. Not tutorials — verified references for AI agents and engineers.

Build system for AI coding-agent context. Generate deterministic, explainable context artifacts for Claude, Codex, Cursor, and MCP.

Multi-domain command-and-control platform for autonomous robotic systems — air, ground, maritime, cyber, space. Open-core C2 with real-time simulation, sensor fusion, mesh networking, and extensible plugin system.

The open-source, local-first alternative to LangSmith. No cloud. No setup.

Academic paper downloader skill: Scopus search + Unpaywall primary + optional Sci-Hub fallback

[ICML 2026] This is the official implementation for paper HiPER: Hierarchical Reinforcement Learning with Explicit Credit Assignment for Large Language Model Agents

Self-hosted autonomous AI agent for Linux — multi-LLM, real desktop control (VNC), WhatsApp, RAG knowledge base, sandboxed security layer, OpenClaw skills

Transform your AI Agent into a PlayStation 2 Reverse Engineering God. An autonomous Claude Code skill packing the MIPS R5900 ISA, topological multimodal vision, and the complete PS2Recomp knowledge.

Agentic SWMM is an automated, auditable, and memory-informed framework for reproducible stormwater modelling, integrating QGIS and EPA SWMM through the aiswmm runtime, reusable Skills, and MCP interfaces, with QA verification, provenance tracking, calibration support, and Codex, Hermes, Claude code as well as OpenClaw compatibility

Local MCP server that exposes classic Outlook desktop (mail, calendar, contacts, tasks, rules) to Claude and other MCP clients via COM. No Azure app, no Graph API, no OAuth.

Claude Code marketplace for multi-agent collaboration — task splitter, output reconciler, adversarial debate, shared memory, acceptance gate. Composes with codex-delegate / gemini-delegate.

Memory staleness detection and auto-expiry for LLM agents

Agent_Console（智能体控制台）是一个本地运行的 Agent + RAG 工作台。它把文档入库、混合检索、查询改写、工具调用、运行回调和人工审核放在同一个控制台里，适合做企业知识库问答、课程资料检索、客服辅助和可追踪的智能体实验

Private, self-hosted document chat for attorneys: parse legal PDFs and query them with local open-source LLMs (Ollama) + verifiable page citations. One-click desktop app at docuchat.app.

Claude Code skill that applies Naur's Programming as Theory Building to coding-agent workflows.

A monorepo for creating, managing, and publishing reusable Agent skills (Claude Code skills).

Terminal UI toolkit for agentic and chat apps, built on rich + prompt_toolkit.

IELTS listening, speaking, reading and writing modules practice and evaluation with IELTS band calculation based on speech and text analysis and evaluation.

面向个人投资者金融投资研究与辅助决策agent系统

MemSearcher is a search agent that keeps a compact, iteratively-updated memory instead of the full interaction history, trained end-to-end with multi-context GRPO.

Research Math Agents, Official code release for our paper RMA

AgentAPI is a lightweight Python framework for building agentic AI backends and chatbots with minimal configuration. It provides a plug‑and‑play developer experience for integrating multiple LLM providers, creating AI agents, streaming responses, managing tools, and handling conversation memory --- all on top of FastAPI.

Single source of truth for GenAI and agentic AI security incidents, mapped to OWASP LLM Top 10, OWASP Agentic Top 10 (ASI), NIST AI RMF, and MITRE ATLAS.

A collection of CLI tools and skills designed to be useful for LLM agents.

Auto-evolving LLM Agent Harness - Benchmark-driven evolution via Claude Code + self_evolution.md guide

Massive multi-source academic literature search for Claude Code — one skill fans out subagents across 20+ scholarly databases (arXiv, Semantic Scholar, Crossref, OpenAlex, PubMed, …), merges into a deduplicated ranked corpus, and acquires the original PDFs.

Local NotebookLM for Claude Code via Google Antigravity (agy / Gemini 3.x): /agy:notebook turns a folder of documents into per-doc summaries + a relevance index + a cited synthesis + Q&A. Plus audio/video transcription, deep web research with citations & branded HTML reports. 13 commands, no Node runtime.

An autonomous agent for open-source repository management — triages issues, reviews PRs across iterations, watches your sites, and escalates only the decisions that are yours

Compile an AI agent's repeated workflows into deterministic, auditable routines that replay for free, with a fallback to the agent.

Self-hosted AI agent that builds apps. Chat with a coding agent (Claude Code or Codex) that builds mini-apps, modifies its own UI, and gets sharper the more you use it. Your personal AI operating system, on your own server.

面向 Codex、Claude Code 与 AI Agent 的考研择校导航 Skill：联网调研复试线、统考名额、推免和拟录取数据，生成避坑提示与冲稳保建议。

📈 Claude Agent Skills for A股分析 | 29个即插即用 Skills 覆盖数据采集/大盘分析/资金流向/涨停追踪/技术面/基本面/估值/财报/多因子/回测/风控/智能报告 | 含自选股监控 | akshare+tushare+东财 | Claude Code 首选搭档

Open-source test harness for AI agents. Stress-test production agents with adversarial multi-turn scenarios in CI

MCP server for parametric 3D modeling in Autodesk Inventor - drive Inventor with AI in natural language. Includes Agent Skills and undocumented Inventor 2026 COM API knowledge.

Ask your video library anything — an AI agent that watches, reasons in a tool-use loop, and answers with playable clips and charts.

gradio mcp-server region:us

gradio region:us

gradio region:us

gradio region:us

docker streamlit region:us

docker agent-collab region:us

streamlit region:us

transformers pytorch text-generation custom_code en arxiv:2312.13286

transformers pytorch gguf codeshell text-generation wisdomshell

transformers gguf llama text-generation llama-factory orpo

transformers pytorch tf jax gpt2 text-generation

gradio region:us

gradio agent-demo-track mcp-server region:us

gradio mcp-server-track mcp-server region:us

transformers safetensors internlm feature-extraction chemistry text-generation

transformers safetensors qwen3_moe text-generation unsloth conversational

transformers safetensors qwen2 text-generation code codeqwen

gguf text-generation base_model:Qwen/Qwen3-Coder-Next base_model:quantized:Qwen/Qwen3-Coder-Next license:apache-2.0 endpoints_compatible

transformers safetensors qwen3_5_moe image-text-to-text moe vlm

transformers safetensors qwen2 text-generation code codeqwen

Framework for building sophisticated Investment Research AI agents .

基于Multi-Agent架构的智能旅游助手

Local-first multi-agent market research desktop workspace built with Tauri, Next.js, and an embedded TradingAgents-compatible core.

Academic advisor due-diligence skill for Codex, Claude Code, and other agents | 学术导师调查评估

MM-ToolSandBox: A Unified Framework for Evaluating Visual Tool-Calling Agents

A self-improving swarm of local-LLM agents that mine, smelt, build, farm, and fight their way through Minecraft as a coordinated team. Built on mineflayer + Ollama.

Reproducing AgenticQwen (arXiv:2604.21590) — dual-flywheel data synthesis + GRPO RL training for agentic small LLMs

TRACE is a practical agentic jailbreak framework that leverages task decomposition, task-aware disguising scenarios and adaptive self-evolution to investigate emerging security risks of LLM-based agents.

基于claude sdk的自动化小说撰写Agent，支持对话，高质量的小说写作与发表

Ship AI-agent artifacts, not demo theater — verifier-gated, cross-family runs that learn which model wins. TRUE recursive self-improvement: its apply loop promotes changes to its own prompts and workflows with no human approval — real measurement required, nothing reviewed first. Run on a throwaway worktree; see docs/SAFETY.md.

Your calls, handled by AI — open-source AI phone agent on a Quectel EC20/EG25 4G modem. Auto-answers calls with realtime voice AI (Qwen/OpenAI/Doubao), dials out, sends SMS, navigates IVR menus.

PAST-Bench: Benchmarking the Foundations of Recursive Self-Improvement in Personal Agents

TRACE Bench: Task-driven Roleplay Agentic Checklist Evaluation

Local browser agent: MLX + Chrome DevTools Protocol. Handles cross-origin iframes, Shadow DOM, ProseMirror. Runs on Apple Silicon, no cloud.

一个 agent skill：把一个主题做成封面/目录/每章插画/精排俱全、可直接出片的 PDF 书。多 agent 调研→逐章写作→配图→Typst 排版→事实核查。工具无关(Claude Code / Codex)、风格可插拔

Autonomous offensive-security agent: recon, authorized pentest, vuln hunting, DFIR & multi-framework compliance from a single prompt. Local-first (llama.cpp), skill-based, deterministic-first.

企业级多智能体AI管理系统 | 企业级智慧管理解决方案 | 基于MCP架构的财务分析、成本预测、知识管理、效能评估一体化平台|- 多智能体系统 (Multi-agent System) - MCP协议 (Model Context Protocol) - - AI决策支持系统 - 企业级AI应用 - Python + FastAPI + Streamlit

Agente autónomo para LALIGA Fantasy: lee el juego, decide y actúa (pujas, ventas, clausulazos, alineación). Python stdlib + agente LLM opcional (Hermes) + UI de supervisión.

Use this cross-platform skill in Codex or Claude Code to establish repository-local continuity memory so a future agent can recover objective, status, decisions, validation, risks, and next actions without relying on previous chat history.

面向"面试官 × 简历主人公"场景的个人简历 RAG 问答 Agent：把简历、证书、荣誉、项目介绍放进知识库，面试官可对简历任意提问（自我介绍、项目深挖、技术八股、HR 素质、简历细节），系统以第一人称自然作答。核心设计是"宽松推理 + 诚实标注"——硬事实（数字/日期/证书名）必须来自检索，证据不足时强制标注"根据现有知识库推测"而非编造。支持多轮追问记忆、检索兜底链、置信度分级回答、访问码闸门与 IP 限流防刷。

Runtime security for AI agents. In-process, zero dependencies, Apache 2.0.

The agentic debugger for Claude Code, Codex CLI and Gemini CLI: a token usage, cost and context-window tracker that shows what the tokens ARE — live memory map, per-file tokens, cache economics, compaction autopsies, PDF reports.

Red-team your RAG pipeline for prompt injection, source-document leakage, and cross-document smuggling, in CI. The gap between RAG eval (RAGAS/DeepEval) and LLM scanners (garak).

Multi-language Active Inference framework combining ant-colony-optimization-inspired multi-agent simulation with variational message passing, implemented across 50 programming languages for research and benchmarking.

Type-aware, evidence-backed delivery contract for AI coding harnesses. Different agents in. One predictable human handoff out.

Phase-based AI development framework for AI Coding Agents. 17 specialized agents across 6 phases, handing off through files in your repo — and it reads the codebase you already have.

Autonomous multi-market trading agent for Hyperliquid

AI-assisted validation toolkit for water, wastewater, and stormwater Utility Network schema mappings.

Python desktop AI assistant for screen awareness, voice interaction, persistent memory, and computer automation

Autonomous agent loops for Hermes Agent: the autoresearch harness and the health/improvement maintenance loops.

The reliability & architecture analyzer for LLM and agentic systems. By ActaClad.

Automated faceless YouTube video pipeline - clone any channel's style, generate videos on autopilot with Claude Code.

Claude Design in Claude Code, Codex, Cursor, and 20+ agents. Designs come from your real components, tokens, and states; approved revisions sync back.

Turn an approved product design, PRD, prototype, design system, or architecture brief into an AI-governed, deployable full-stack project with a pnpm monorepo, requirement traceability, canonical multi-agent rules, mandatory AI review, executable quality gates, CI, container deployment assets, and handoff evidence.