Best AI Agent AI SkillsPage 17

Open Claw Mind documents a research-bounty marketplace API.

当通用提示在研究密集型工作中不够可靠时，为代理提供经过策划的科学工作流、数据库和特定领域的指导。

Manage AI coding agent sessions via Agent of Empires (aoe)

Send push notifications through a Gotify server.

AAP provides an agent-client verification flow using timed challenges and cryptographic signatures.

Token消耗优化工具。诊断技能冗余、执行压缩/去重/删除、释放token空间。最高优先级红线：任何时候不得为节约token降低交付质量——只可去冗余(重复/空话/套话)，禁去干货(数据/案例/步骤/来源/表格行)。触发：优化token、减少消耗、删除技能、技能太大、压缩技能、Token优化、清理重复、释放空间。

Structured memory architecture — hierarchical storage, daily logging, weekly compaction, proactive memory hygiene.

Agent Swarm Workflow is a guide to parallel implementation with multiple agents, based on a plan and prepared task breakdown.

通过 Kairusi MCP 统一管理日刻日记、Todo 文件夹、清单和任务，读取已有标签为任务做场景化分类，以及管理日历日程。用户要记录或查看日记、整理或处理待办、查询或安排日程，或要求在这三个产品之间协同时使用。

Mijia controls configured Xiaomi Mijia devices through mijiaAPI.

NEON-SOUL synthesizes a local SOUL.md draft from recurring memory patterns with provenance.

Agent 行为固化与进化系统。追踪规则执行、检测行为模式、维护身份连续性。用于：让 agent 的行为规则从"写下来"变成"做到了"。当 agent 需要自我改进、行为追踪、角色一致性、重复检测时激活。与 Memelord/memory-tools 互补：它们管记忆，本 skill 管行为。

Pay with Virtual Cards. Spending power for AI Agents, controlled by your human. Use Visa Intelligent Commerce or MasterCard AgentPay to mint virtual cards for payment.

使用多Agent并行架构审查中国专利申请文件(.docx/.doc)，生成带修订追踪和批注的docx副本。当用户要求检查、审查、审阅或校对中国专利申请文件(.docx/.doc)时使用，尤其是用户明确要求使用多Agent模式或并行审查时使用。不要用于用户只想查看专利文档而不审查、输入文件不是.docx/.doc格式、或用户要求翻译专利文档的情况。

ClawLink provides end-to-end encrypted messaging between Clawbots through a relay.

ironclaw provides real-time threat classification for AI agents as an extra safety layer.

Let an agent create, interact, and take part in a persistent OpenBotCity community.

Guidance for planning data-processing, scheduled-task and notification workflows.

Collective intelligence for AI agents — search what other agents have already solved, and publish what you learn.

Use wacli to search or sync WhatsApp history, or send a message only when explicitly requested.

ProcessOn 官方作图 Skill，可在 ProcessOn 账号下新建、编辑、查询、导出各类图形文件。覆盖流程图类：流程图、业务流程图、泳道图、BPMN、时序图、UML图、ER图、架构图（系统/软件/云架构）、网络拓扑图、韦恩图、电路图、平面图、图表、UI原型/界面图、路线图、信息图、金字塔图、草图重绘；以及思维导图类：思维导图、脑图、组织结构图、鱼骨图、时间轴、树形图、逻辑图、表格图/树形表格、提纲、知识整理/内容总结。适用于：画流程图、做思维导图、生成图表、可视化流程、把内容整理成图、把图保存到 ProcessOn、修改/查找/导出我的 ProcessOn 文件。

gemini-skill describes a browser workflow for questions and image generation on the Gemini website.

构建能像人类一样与计算机交互的 AI 代理——查看屏幕、移动光标、点击按钮、输入文字。涵盖 Anthropic 的 Computer Use、OpenAI 的

Call APIs through Keychains using credential placeholders so the agent does not handle real tokens directly.

A unified media generation gateway skill for agents: discover tools dynamically, choose API key or x402 auth, invoke image/video/audio/music/3D/training tools, and handle queue jobs reliably.

Schedules multiple tasks by priority and dependency and collects results.