AI Agents not only answer questions, but can also break down targets, call external tools, and perform multiple steps in succession. This page aggregates agent products suitable for research, operations, development, and personal efficiency scenarios, comparing their autonomy, tool permissions, memory capabilities, manual verification mechanisms, and traceability after task failure.
Hermes Agent is a master agent framework launched by Nous Research, Hermes Agent runs on your server, has persistent memory and self-learning loops, and can automatically precipitate practical task experience into reusable skills, so that it becomes stronger and stronger. Hermes Agent supports multi-platform access (such as CLI, email, and multiple messaging channels) for long-running automation assistants and team workflows. To improve controllability, Hermes Agent provides a real sandbox and multiple execution backends (local, Docker, SSH, etc.), and combines permission confirmation and isolation mechanisms to make tool calls more secure. Whether it's for code tasks, data organization, or scheduled automation, Hermes Agent serves as a pluggable capability base in the MCP ecosystem, continuously expanding your agent skill base.
Agent-Skills.md is an Agent Skills Directory and Market for AI Agents. It mainly uses the unified SKILL.md specification to precipitate and distribute reusable capability packs. Agent-Skills.md indexes various skill warehouses and provides search, tag filtering, preview and download, allowing developers to quickly find skills suitable for Claude Code, Cursor, Gemini CLI, OpenAI Codex and other tools. The platform emphasizes the "production-available" skill structure and metadata specification, which facilitates the reuse of tool invocation, workflow orchestration, code review, automated operation and maintenance, and office processes in the MCP ecosystem in a modular manner. With the help of submission portal and continuous inclusion mechanism, Agent-Skills.md makes it easier for teams to build a manageable and scalable skill base and improve the efficiency and consistency of agent implementation.
Skills Directory is an Agent Skills Directory focusing on safety and quality. Skills Directory provides an "available, verifiable, installable" set of skills for Claude and related agent scenarios. Skills Directory focuses on security first. It conducts automated security analysis and rule detection for each skill, focusing on identifying risks such as prompt word injection, credential theft, data dissemination, and malicious scripts, and helps users quickly filter trusted skills based on level and status. Through classified browsing, search, and one-click copy/installation methods, Skills Directory makes it easier to reuse and implement workflows such as code review, test generation, document writing, and research retrieval. For teams, Skills Directory is also suitable as a skills management portal, taking into account security compliance and stability when expanding agent capabilities.
Agent Skills (agentskills.me) is an Agent Skills Skills Directory and Installation Navigation Station for AI Agents. It focuses on a reusable skill set based on the SKILL.md specification. Agent Skills supports the discovery and selection of skills for Claude Code, Cursor, OpenCode, Gemini CLI, Codex CLI and other platforms, and provides clear installation guidelines, such as copying SKILL.md to a specified skills folder or one-click installation through the command line. Through classified browsing, skill details pages and download entrances, Agent Skills helps developers manage skill packs more efficiently and expand agent capabilities. It is suitable for building pluggable workflow and automation capabilities stacks under the MCP ecosystem.
Skillstore is an AI skills market for Claude, Codex and Claude Code, focusing on the distribution of reusable skills with "security audit + one-click installation". Skillstore accumulates skills in the form of standardized packages through automated security scanning and versioning management, making it convenient for developers to quickly discover, install and share Agent Skills for code review, test generation, refactoring optimization, document writing and security inspection. High-frequency scenarios. Skillstore also supports submitting skills and triggering security audit and audit processes, allowing teams to build skills libraries and skill stacks more controllably, improve the reliability and maintainability of agent expansion capabilities, and adapt to pluggable capabilities in the MCP ecosystem.
Skills.sh is an open Agent Skills catalog and distribution platform for AI agents. Skills.sh aggregates reusable skills and allows developers to quickly install "executable process knowledge" for agents with one command. Through the installation method of npx skills add, Skills.sh publishes and manages skills in a standardized structure, which is convenient for unified access in multiple agent environments such as Claude Code, Cursor, GitHub Copilot, and Gemini. Skills.sh provides skills search, leaderboards and trend views to help you select high-quality skills by scenario for tasks such as code specification, automated processes, document generation and tool integration, achieving more controllable capability expansion and team reuse.
AI Templates (aitmpl.com) is a catalog of AI programming templates and components for developer agents such as Claude Code, which centrally organizes directly reusable Agents, Commands, Skills, and MCP integration solutions. AI Templates provide search and classified browsing capabilities to help you quickly find common AI programming components such as code review, test generation, scaffolding, and automated workflow, and combine selected components into an installable skill stack through "Stack Builder". For teams that want to engineer agent capabilities, AI Templates takes into account component discovery, combination and integration efficiency, making Claude Code templates, MCP connections and skill reuse smoother, and improving the quality and speed of daily development and delivery.
Molili is an AI agent desktop application that can truly do things. It is mainly upgraded from "dialogue response" to "automatic execution", handing over repeated tasks such as browsing the web, filling out forms, reading and writing files, and organizing data to Molili. Molili is installed and runs locally and can directly access real files and applications, making it more suitable for daily AI office and personal efficiency improvement. You can also issue instructions remotely through WeChat, Dingnail, Flying Book and even Siri, allowing Molili to continue executing 7×24 hours a day and synchronize feedback on progress and results. In conjunction with the expanding skills ecosystem, Molili can cover high-frequency scenarios such as office automation, information collection, and document processing, making Molili easier to use and more efficient.
CocoLoop Skills Store is a faster and safer AI Agent Skills Market launched by CocoLoop. It provides a searchable, downloadable, and one-click installable skill catalog for agents such as OpenClaw and Molili. The CocoLoop Skill store highlights AI compliance and security governance capabilities, provides security ratings, permissions and risk warnings for each skill, and combines dimensions such as source credibility, executable content review, and privacy statements to help users determine security boundaries before installation. Through unified installation commands and ZIP package distribution, the CocoLoop Skill Store makes skills access lighter and suitable for multiple scenarios such as office automation, content processing, and development assistance, reducing skills trial costs and improving the controllability and controllability of skills use. Compliance.
SkillsMP is an Agent Skills Market and Aggregate Directory for AI Programming Assistants. It mainly collects and organizes a large number of Agent Skills from GitHub to help developers discover, filter and install reusable skills faster. SkillsMP opens standards around SKILL.md and is compatible with tools such as Claude Code, OpenAI Codex CLI, and ChatGPT, allowing skills to precipitate best practices, workflow, and code specifications in a unified structure. Through classified navigation, author and popularity screening, quality indicators and search capabilities, SkillsMP makes it easier to reuse and combine skills such as code review, test generation, automated submission, and project scaffolding in AI programming scenarios. SkillsMP also provides documentation and interface instructions to facilitate teams to integrate skill management into engineering processes and continuously improve AI programming efficiency and delivery quality.
Agent Skills Catalog is an AI agent skill catalog and governance platform launched by The Trust Registry, focusing on providing "security-certified" installable skills for agents such as Claude, Cursor, and Copilot. Agent Skills Catalog unifies management of skills from development, testing to release: supports generation and debugging skills, conducts consistency verification in multiple agent environments, and lists qualified skills in public catalogs or corporate private catalogs. The platform emphasizes AI compliance and security audit capabilities, including vulnerability scanning, key and sensitive information detection, dangerous pattern identification, and specification verification of Agent Skills standards, helping teams expand agent capabilities in a more controllable way and reduce online risks. and improve skills reuse efficiency.
ClawHub is a skills registration and distribution platform for AI agents. It focuses on a "versioned skill warehouse", allowing developers to manage skill packages like managing dependencies. ClawHub supports browsing, installing and publishing skill packs, and provides a searchable skill catalog and popular selections to quickly find suitable agent capability modules. ClawHub emphasizes that versions can be rolled back and stably iterated, and cooperates with the command-line one-click installation process to reduce skills access costs and improve the usability of agents in automated tasks, tool calls, and workflow orchestration. For building a pluggable capability system based on MCP, ClawHub can also serve as an entrance to the skills ecosystem, helping teams reuse and govern AI agent skills more efficiently.
Skill Hub China is an Agent Skill ecosystem platform that integrates practical Skill cases and reusable solutions, helping developers and product teams quickly find skills and workflows that are useful, easy to use and reusable. Skill Hub China provides Skill navigation, practical article aggregation and introductory tutorials around AI programming, AI office and automation scenarios, allowing you to implement Agent Skills into real tasks at a lower cost. Through mechanisms such as "submitting skills, sharing practices, and community co-creation", Skill Hub China centralizes the management of scattered open source skills and best practices, which is easy to retrieve, reuse and continuous iteration. It is suitable for building a pluggable capability system based on MCP. and productivity tool chain.
Tencent SkillHub is an AI Skills skills community and distribution platform optimized for China users, helping agents quickly discover, download and manage reusable skills. Tencent SkillHub provides Chinese search and a selection of high-quality skills sets, covering scenarios such as office automation, development tool chains, data processing and knowledge organization, making AI assistants more stable and efficient through skill expansion. Through the accelerated and unified installation process of domestic nodes, Tencent SkillHub lowers the threshold for skill access and supports standardized handling of skill sources and copyright issues. It is suitable for building a pluggable agent capability system and MCP ecosystem integration.
360 Security Skills Center SkillHub is a trusted Skills Center and secure source repository for AI assistants, providing high-quality skills that can be discovered, shared, and installed. SkillHub combines static content auditing with dynamic sandbox verification to conduct security scanning and risk interception of skill codes, and provides hashed tamper-proof verification to help AI assistants be safer and more controllable when installing and running skills. Whether it is 360 Security Lobster, OpenClaw or other agents, SkillHub can be used as a security technology energy source to achieve domestic accelerated downloads, cloud filtering and trusted skills management.
ZeroClaw is an AI chat assistant that focuses on local operation and privacy protection, emphasizing data is native, lightweight to deploy, and safe to use. ZeroClaw runs on users' own devices and can connect to communication channels such as Telegram, Discord, WhatsApp, etc., making it suitable for daily Q&A, personal assistants, message processing, and automated interactions. For users who value data security and local control, ZeroClaw is not just an ordinary conversation tool, but also an AI chat product that takes into account privatization deployment, continuous memory, and task execution capabilities, improving communication and collaboration efficiency while ensuring privacy.
NVIDIA NemoClaw is an open-source AI office product launched by NVIDIA, focusing on adding privacy and security controls to OpenClaw, allowing users to deploy intelligent assistants that are online 24×7 hours a day with more peace of mind. NVIDIA NemoClaw supports running autonomous agents in on-premises or dedicated computing environments, combined with policy-based privacy protection, security guardrails, and high-performance open-source model capabilities, making it ideal for information organization, content processing, task execution, and continuous office collaboration. For individuals and teams concerned about data security, deployment flexibility, and on-premises experience, NVIDIA NemoClaw is not just an intelligent assistant framework, but an AI office tool that balances execution, privacy, and efficiency.
HiClaw is an open-source AI office product launched by Higress, focusing on multi-agent collaboration and human supervision based on instant messaging protocols. HiClaw coordinates multiple Worker Agents to complete complex tasks through Manager Agent, which is suitable for scenarios such as information organization, task distribution, process collaboration, and multi-step work execution. As an AI office tool that emphasizes control and collaboration, HiClaw supports creating tasks in conversations, monitoring progress, and integrating AI gateways with MCP tools to expand real-world execution capabilities. For teams that need to balance automation efficiency, security management, and manual intervention, HiClaw helps users transform complex office processes into clearer and more efficient intelligent collaboration systems.
PoorClaw is an out-of-the-box free AI chat tool that focuses on connecting to mainstream AI models with one click, allowing users to quickly obtain a stable intelligent conversation experience without the need for API keys or complex configurations. PoorClaw supports direct desktop use and can also be connected to various channels such as Telegram and QQ Bot, making it suitable for daily Q&A, information acquisition, content generation, and multi-scenario intelligent interaction. For individual users who want to experience intelligent assistants at a low cost, PoorClaw is not only a simple chat entrance, but also an AI chat product that takes into account convenient deployment, multi-model support, and scalability capabilities, which can help users more easily access the OpenClaw ecosystem and mainstream large model capabilities.
LinkFox Claw is an AI office assistant for cross-border e-commerce scenarios, focusing on improving operational efficiency on platforms such as Amazon, Shopify, TikTok, and Temu. LinkFox Claw mainly serves high-frequency services such as product selection analysis, market research, data collation, and operation execution, helping cross-border e-commerce teams complete complex information processing and daily operations more efficiently. For sellers and operators who need to improve the efficiency of cross-border business, LinkFox Claw is not just an ordinary conversation tool, but also an AI office product that combines e-commerce data, research capabilities, and operational assistance capabilities, suitable for daily store operations, product research, and business decision support.
CoPaw is an AI office assistant launched by the AgentScope team, positioned as a personal intelligent assistant that understands your needs and is by your side. CoPaw supports simplified installation, can be deployed on-premises and in the cloud, and provides multi-terminal access and capability expansion, making it suitable for daily office, information organization, content processing, and task collaboration. As an AI office product for real workflows, CoPaw not only supports interactions in chat software, but also has capabilities such as skill expansion, long-term memory, and console management, helping individuals and teams transform natural language instructions into more efficient office execution processes, improving continuous collaboration and automated processing efficiency.
EasyClaw is a native desktop AI office assistant built on the OpenClaw framework, supporting one-click installation and zero-configuration use on Mac and Windows, and can be quickly used without API Key. EasyClaw focuses on AI office and task automation scenarios, and can operate the interface and web applications like a real person, helping users complete tasks such as document summarization, data cleaning, email processing, price monitoring, code review, and information organization. EasyClaw also supports a cloud version with multi-channel messaging connections, making it suitable for users who need 24×7 hours of online collaboration. For individuals and teams looking to translate natural language instructions directly into desktop executions, EasyClaw is an AI office tool that combines local control, security, and efficiency.
Wukong is an AI office product launched by DingTalk, focusing on moving from conversational AI to mobile AI, not only answering questions or generating content, but also directly operating computers to complete actual tasks. Wukong focuses on intelligent office, task execution, and process collaboration scenarios, making it suitable for document processing, information organization, workflow promotion, and multi-step office automation. For individuals and teams looking to improve efficiency, Wukong is not only an intelligent assistant, but also an AI office tool that can truly participate in execution, continuous collaboration, and integration into the enterprise office system, helping users complete complex tasks more efficiently.
AstronClaw is an AI office assistant launched by iFLYTEK, focusing on the rapid deployment of exclusive AI assistants and 24×7 hours online services. Based on OpenClaw's core capabilities, AstronClaw supports the creation of deeply customizable personal AI assistants and can integrate efficient skills to enable multi-channel information interaction. As an AI office product for efficiency improvement and task collaboration scenarios, AstronClaw is suitable for information processing, content generation, office communication, and daily workflow collaboration, helping individuals and teams achieve a continuous online and scalable smart office experience with a lower threshold.