GitHub stars favor AI agents designed for interaction and productivity, including NousResearch’s hermes-agent and affaan-m’s ECC for optimizing agent skills. Token compression to reduce LLM costs leaps with chopratejas/headroom. New tools for gathering and synthesizing information from social and web platforms like mvanhorn/last30days-skill and Panniantong/Agent-Reach gain traction. PaddlePaddle/PaddleOCR boosts document understanding with powerful OCR supporting 100+ languages. NVIDIA’s cosmos extends AI into physical robotics and environments.

NousResearch/hermes-agent

⭐ 183,518 stars · +1845 today · Python · topics: ai, ai-agent, ai-agents, anthropic, chatgpt, claude, claude-code, clawdbot, codex, hermes, hermes-agent, llm, moltbot, nous-research, openai, openclaw

An evolving AI agent platform that grows capabilities over time, blending insights from Claude and OpenAI models to support advanced conversational agents.

chopratejas/headroom

⭐ 14,700 stars · +2473 today · Python · topics: agent, ai, anthropic, claude-code, compression, context-engineering, context-window, cursor, fastapi, langchain, llm, mcp, openai, prompt-engineering, proxy, python, rag, token-optimization, tokens, typescript

A compression tool that reduces token counts by up to 95% before they reach large language models, cutting API costs while preserving output accuracy.

CopilotKit/CopilotKit

⭐ 32,789 stars · +366 today · TypeScript · topics: agent, agent-native, agentic-ai, agents, ai, ai-agent, ai-assistant, assistant, assistant-chat-bots, copilot, copilot-chat, generative-ui, js, llm, nextjs, open-source, react, reactjs, ts, typescript

Frontend tools and protocols for building agent-driven user interfaces in React and Angular, facilitating generative AI assistant deployment.

lfnovo/open-notebook

⭐ 26,135 stars · +1152 today · TypeScript · topics: assistant, learning, note-taking, notebook, notes-app, self-learning

An open source version of Notebook LM focusing on flexible and feature-rich note-taking to support learning and information management.

affaan-m/ECC

⭐ 208,506 stars · +1361 today · JavaScript · topics: ai-agents, anthropic, claude, claude-code, developer-tools, llm, mcp, productivity

A performance optimization framework for AI agents emphasizing skills, memory, and security enhancements across Claude Code and related platforms.

Panniantong/Agent-Reach

⭐ 21,710 stars · +148 today · Python · topics: agent-infrastructure, ai-agent, ai-search, automation, bilibili, claude-code, cli, cursor, free-api, llm-tools, mcp, python, reddit-scraper, twitter-scraper, web-scraper, xiaohongshu, youtube-transcript

CLI tool enabling AI agents to crawl and search across multiple web platforms like Twitter, Reddit, YouTube, and Chinese apps without API fees.

NVIDIA/cosmos

⭐ 9,468 stars · +479 today · Jupyter Notebook

Open platform offering world models and datasets to build AI for robotics, autonomous vehicles, and smart infrastructure applications.

666ghj/MiroFish

⭐ 64,798 stars · +320 today · Python · topics: agent-memory, financial-forecasting, future-prediction, knowledge-graph, llms, multi-agent-simulation, public-opinion-analysis, python3, social-prediction, swarm-intelligence

Swarm intelligence engine that predicts outcomes by simulating collective behavior, useful for financial forecasting and social opinion analysis.

mvanhorn/last30days-skill

⭐ 28,296 stars · +731 today · Python · topics: ai-prompts, ai-skill, bluesky, claude, claude-code, clawhub, deep-research, hackernews, instagram, openclaw, polymarket, recency, reddit, research, social-media, tiktok, trends, twitter, web-search, youtube

AI skill that researches trending topics across social networks and web sources, then generates grounded summaries for deep insights.

PaddlePaddle/PaddleOCR

⭐ 80,622 stars · +747 today · Python · topics: ai4science, chineseocr, document-parsing, document-translation, kie, ocr, paddleocr-vl, pdf-extractor-rag, pdf-parser, pdf2markdown, pp-ocr, pp-structure, rag

Lightweight OCR toolkit converting images and PDFs into structured data with multilingual support, aiding document parsing and LLM pipelines.

Source: GitHub Trending — Daily [1]