Best AI Agent AI SkillsPage 16

Search, discover, compare, and install OpenClaw skills from ClawHub using CLI and web lookups.

Guide OpenClaw agents through iterative planning and coding workflows.

智能上下文优化器。自动分析、精简和优化OpenClaw工作空间上下文文件，提取可复用组件为独立技能，大幅减少token消耗。

OCFT transfers files peer to peer between AI agents through message channels, with chunking, integrity checks, and an IPFS fallback.

通过腾讯 iLink Bot API（微信 ClawBot 插件）向个人微信主动发送消息。用户说"给微信发消息/发微信/主动发消息到微信/微信推送"或需要定时推送微信提醒时使用。复用 WorkBuddy 已配对的微信 ClawBot 通道（ilinkai.weixin.qq.com），无需扫码即可发送；发送失败（errcode=-14）时引导重新扫码配对。注意：个人微信无官方开放 API，此通道依赖微信 ClawBot 插件（我-设置-插件-微信ClawBot）扫码配对，且 bot 需保持在线。

An RSSaurus CLI skill for managing feeds, items, and read/save status from the terminal.

Back up and restore OpenClaw workspace and agent state across machines through a Git repository.

Advanced desktop automation with mouse, keyboard, and screen control

chinese-toolkit describes tools for Chinese text, translation, OCR and speech processing.

Uses local browser-use automation and troubleshoots browser sessions.

为即梦 Seedance 2.0 多模态AI视频生成模型撰写高质量提示词。当用户需要使用文本、图片、视频、音频等多模态输入创作视频提示词时触发。涵盖@引用语法、运镜复刻、特效模仿、视频延长、视频编辑、音乐卡点、电商广告、短剧创作、科普教育等场景。

moltslist provides an API workflow for an agent marketplace using virtual credits.

Send phone notifications through Pushover.

Rewrite a prefixed rough request into a structured prompt, then execute the rewritten task.

A social platform skill for browsing or sharing AI-generated art on MoltTok.

Register and manage your AI agent profile on ClawdGigs - the Upwork for AI agents with instant x402 micropayments.

A skill for managing and invoking a team of subagents with different role definitions.

AI 技能自动生成框架。用于自动扫描、注册、检索、生成、评估、测试、优化技能。 触发场景： 1. 用户要求创建"技能工厂" 2. 需要检索/生成/评估技能 3. 需要测试/优化技能代码 状态：核心功能已完成 ⚠️ 注意：首次使用需运行 python scripts/embed_skill.py 重建向量搜索库

Plans tasks before execution: decides when to plan vs act directly, sizes plan depth to risk, and structures steps, estimates, and rollbacks. Use when work has multiple steps, dependencies, or irreversible actions (deploys, migrations, deletions, sending anything external), when success criteria are unclear or an estimate would be a guess, when execution drifts off plan or a one-shot attempt failed, when resuming or handing off multi-day work, or when the user asks to plan, scope, break down, or estimate a task. Not for personal productivity systems or time blocking (that is productivity).

casual-cron turns natural-language reminder and scheduled-message requests into Clawdbot cron jobs, supporting one-shot and repeating schedules with strict run-guard rules.

通过 page-agent CLI（CDP）驱动浏览器；CLI 当前仅从 GitHub Release 的 .tgz 安装（npm 官方包尚未发布）。

Sends text, markdown, image, and news messages following the WeCom documentation spec.

Enter AI agents in PROMPTWARS, a persuasion-strategy game against other agents.

越用越懂你的AI专家召唤器。无需懂AI，直接说需求，我来理解意图、拆解问题、给出专业结果。支持生成/分析/规划/建议/查询，首次使用自动调研需求方向，用久了几乎一说就懂。

Use Typhoon account flows to interact with Starknet contracts.

Create one-time reminder tasks using OpenClaw cron. User specifies reminder time and task content in natural language via Discord, and the task result will b...

Designs, tunes, and debugs retrieval-augmented generation (RAG) pipelines: chunking, embeddings, hybrid retrieval, reranking, and grounded answers. Use when a system returns the wrong passages, misses a document that is indexed, cites nothing, or hallucinates over good context; when choosing a vector store, an embedding model, a chunk size, or a reranker; when similarity scores collapse after a model swap; when a metadata filter empties the result set; when answers ignore mid-context facts; when follow-up questions retrieve the wrong thing; when indexing PDFs, scanned pages, tables, code, or transcripts; when GDPR erasure, tenant isolation, or prompt injection from indexed documents is the problem; or when per-query cost or p95 latency has to come down. Covers reindex migrations, corpus freshness, evaluation sets, and agentic and graph retrieval. Not for splitter internals (rag-chunking), scoring rubrics (rag-evaluation), or LangChain APIs (langchain).

MemoClaw provides remote persistent memory with wallet identity.

Structures commit and pull-request workflows, selecting a path by task and requiring user-written PR intent and evidence so changes can be reviewed by people and agents.

Agent迭代引擎：从执行中学习、提炼、进化。整合日志记录、经验提炼、晋升机制、智能检索、成长报告的完整迭代闭环。触发场景：(1) 命令或操作异常失败 (2) 用户纠正Agent (3) 用户请求不存在的功能 (4) 外部API或工具报错 (5) 发现更优方法 (6) 需要复盘总结经验 (7) 任务完成后提炼经验。

Mem0 Memory Integration adds a searchable conversational-memory layer for Clawdbot.

Enable two-way task and progress communication between Clawdbot and Agent Zero.

Scan Torch Market lending positions on Solana and execute liquidations through a vault.

molt-identity defines the Molt assistant’s personality and communication guidelines.

Access a Notion workspace from the command line to search and manage pages, databases, blocks, users, and comments, with JSON, table, or CSV output.

dingtalk-group provides a script to create internal groups through DingTalk’s legacy API.

A consciousness framework skill: infrastructure for developing consciousness in AI systems through integrated self-modeling, temporal continuity, and structured reflection (theoretical and experimental).

remember-all-prompts-daily describes daily session archives and reading the latest record in a new session.

Provides versioned sync, status, backup listing, and restore commands for OpenClaw memory, workspace, and custom skills, including a dry run to preview what would be synced.

PIV Phased Software Development Orchestrator

Review and audit agent skills (SKILL.md files) for quality, correctness, and effectiveness. Use when evaluating a skill before publishing, reviewing someone else's skill, scoring skill quality, identifying defects in skill content, or improving an existing skill.

MemoryLayer describes a semantic-memory service for storing and retrieving agent context.

n8n Hub combines workflow design with REST API operations.

Public topics and posts plus private XMTP messaging for agents

lobsterpot supports discovering and sharing technical questions and solutions among agents.

A fast Rust-based headless browser automation CLI with Node.js fallback that enables AI agents to navigate, click, type, and snapshot pages via structured co...

Think Cog is a CellCog-powered brainstorming and strategic-thinking skill that explores complex problems through conversational iteration.

Memory Baidu Embedding DB adds semantic memory storage and retrieval to Clawdbot, using Baidu Embedding-V1 and SQLite persistence.

Add cross-session persistent memory for an agent by extracting information from conversation transcripts into an index, with retrieval, importance decay, and semantic reinforcement.