Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.
GitHub repository with 11,656 stars and 759 forks.
Language: Python
Topics: agent, ai, anthropic, claude-code, compression, context-engineering, context-window, cursor, fastapi, langchain
Trending score 3.28, activity score 0.05, stars gained +2,887, forks gained +157.
2026-06-04: 11,656 stars and 759 forks.
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.
GitHub repository with 11,656 stars and 759 forks.
Trending score: 3.28; stars gained: +2,887; forks gained: +157.
Language: Python
Topics: agent, ai, anthropic, claude-code, compression, context-engineering
OpenSquilla — Token-Efficient AI Agent with same budget, higher intelligence density
GitHub repository with 2,854 stars and 201 forks.
Trending score: 3.25; stars gained: +86; forks gained: +8.
Language: Python
Topics: agent, ai, ai-agents, deep-learning, foundation-models, llm
Knowhere extracts, parses, and outputs structured chunks ready for AI Agents and RAG.
GitHub repository with 926 stars and 92 forks.
Trending score: 3.18; stars gained: +75; forks gained: +6.
Language: Python
Topics: agent, ai-agents, chromadb, claude, claude-code, cursor
🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。
GitHub repository with 21,141 stars and 2,858 forks.
Trending score: 2.69; stars gained: +18; forks gained: +6.
Language: Python
Topics: llm, ollama, maxkb, knowledgebase, chatbot, langchain
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.
GitHub repository with 70,452 stars and 9,524 forks.
Trending score: 2.08; stars gained: +99; forks gained: +12.
Language: Python
Topics: agent, agentic, agentic-framework, agentic-workflow, ai, ai-agents
Modular SenseNova skills for building AI-powered office assistants and productivity workflows
GitHub repository with 3,380 stars and 243 forks.
Trending score: 1.90; stars gained: +102; forks gained: +7.
Language: Python
Topics: agent, agent-skills, ai-agents, ai-assistant, data-analysis, document-processing
The agent that grows with you
GitHub repository with 180,526 stars and 30,962 forks.
Trending score: 5.79; stars gained: +1,360; forks gained: +322.
Language: Python
Topics: ai, ai-agent, ai-agents, anthropic, chatgpt, claude
Use claude-code for free in the terminal, VSCode extension or discord like OpenClaw (voice supported)
GitHub repository with 32,285 stars and 4,901 forks.
Trending score: 4.60; stars gained: +456; forks gained: +73.
Language: Python
SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.
GitHub repository with 4,867 stars and 482 forks.
Trending score: 4.55; stars gained: +340; forks gained: +27.
Language: Python
Topics: agent-skills, self-evolving-agents
754 structured cybersecurity skills for AI agents · Mapped to 5 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platforms · 26 security domains · Apache 2.0
GitHub repository with 13,233 stars and 1,551 forks.
Trending score: 4.53; stars gained: +301; forks gained: +38.
Language: Python
Topics: ai-agents, claude-code, cybersecurity, incident-response, mitre-attack, penetration-testing
Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work.
GitHub repository with 4,141 stars and 516 forks.
Trending score: 4.43; stars gained: +415; forks gained: +37.
Language: Python
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
GitHub repository with 130,099 stars and 21,139 forks.
Trending score: 4.42; stars gained: +277; forks gained: +38.
Language: Python
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
GitHub repository with 78,180 stars and 15,369 forks.
Trending score: 3.74; stars gained: +42; forks gained: +12.
Language: TypeScript
Topics: agent, agent-collaboration, agent-harness, ai, cao, chatgpt
Web dashboard for Hermes Agent — multi-platform AI chat, session management, scheduled jobs, usage analytics
GitHub repository with 7,278 stars and 909 forks.
Trending score: 3.74; stars gained: +133; forks gained: +15.
Language: TypeScript
Topics: agent, ai-agent, chat-ui, dashboard, hermes, hermes-agent
Production-ready platform for agentic workflow development.
GitHub repository with 143,856 stars and 22,635 forks.
Trending score: 3.73; stars gained: +125; forks gained: +19.
Language: TypeScript
Topics: agent, agentic-ai, agentic-framework, agentic-workflow, ai, automation
A slide framework built for agents.
GitHub repository with 4,716 stars and 322 forks.
Trending score: 3.62; stars gained: +108; forks gained: +12.
Language: TypeScript
Topics: agent, react, slides
MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research · 浏览器里运行的安卓模拟器 · Browser-hosted Android Simulator · Verifiable Evaluation · Scalable Online RL Training
GitHub repository with 497 stars and 79 forks.
Trending score: 3.43; stars gained: +84; forks gained: +10.
Language: TypeScript
Topics: agent, agents, ai, android, automation, benchmark
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.
GitHub repository with 11,656 stars and 759 forks.
Trending score: 3.28; stars gained: +2,887; forks gained: +157.
Language: Python
Topics: agent, ai, anthropic, claude-code, compression, context-engineering