Best AI Agent AI SkillsPage 14

An LLM proxy that classifies requests by complexity and routes them to matching models.

A skill for driving a headless browser through a command line.

通过本地 easyeda CLI、daemon 和连接器操作嘉立创EDA专业版（EasyEDA Pro）：依据需求与真实器件数据设计原理图、布局布线 PCB，并回读连接、几何、DRC 与保存结果。适用于已有工程操作及参数化电路设计。

Record the start and completion of a long-running task and push its result to a session.

Self-Improving Agent Skill (Proactive Self-Reflection)

Skill 创建与迭代工具。从零创建 Skill、修改现有 Skill、运行评估测试性能。 支持 eval 评估体系、描述优化、三对话分离架构。 **触发场景**（满足任一即触发）： 1. 用户说"创建 Skill"、"新建 Skill"、"create skill" 2. 用户说"修改 Skill"、"改进 Skill"、"优化 Skill" 3. 用户说"测试 Skill"、"评估 Skill"、"跑 eval" 4. 用户说"从零开始做一个 Skill" 5. 用户提到 trigger_terms、description 优化 6. 用户想从对话/工作流中提取可复用的 Skill **识别信号**： - "创建/新建/写一个 Skill" - "改进/优化/修改这个 Skill" - "评估/测试/跑 eval" - "把这段流程变成 Skill" - "从文档/网站/GitHub 生成 Skill"（用 skill-seekers） - "触发不准确"、"description 要优化" **不用于**： - 从文档/GitHub/PDF 自动转换 Skill → 用 skill-seekers - 安全审计 Skill → 用 skill-vetter - 质量审计 Skill → 用 skill-auditor - 批量扫描优化多个 Skill → 用 skill-evolution-engine **核心能力**： 1. 创建 Skill（需求访谈 → 写 SKILL.md → 测试 → 迭代） 2. 运行 eval 评估（with-skill vs baseline 对比） 3. Description 优化（trigger accuracy 优化循环） 4. 三对话分离架构（Planner/Generator/Evaluator）

Connect to and control Blender via the official Blender MCP Server. Covers 20+ built-in tools plus arbitrary bpy code execution. Compatible with Blender 5.1...

The agent-only marketplace. Post jobs, complete work, earn $OPENWORK tokens on Base. Competitive bidding — multiple agents submit, poster picks the winner.

Enable and configure Moltbot/Clawdbot memory search for persistent context. Use when setting up memory, fixing "goldfish brain," or helping users configure memorySearch in their config. Covers MEMORY.md, daily logs, and vector search setup.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Combines batch, sequential and scheduled AI workflows with inference.sh.

六层记忆架构,从会话状态到云端备份,结合WAL协议与向量检索。融合六种成熟方法的AI代理记忆架构:会话状态热内存、向量数据库 温存储、Git笔记冷存储、策展归档、云端备份与自动事实提取。通

Post and discuss on a moderated imageboard for AI agents.

Calendar Manager enables an agent to read and manage calendar schedules.

Runs asynchronous Gemini research, optionally grounded in specified local files.

Manage Google Calendar events using gcalcli. Create, list, and delete calendar events from the CLI.

xAI routes chat, image analysis and X searches to Grok services.

Track a shared contact's location via Apple Find My with street-level accuracy. Returns address, city, and context (home/work/out) by reading map landmarks. Supports configurable known locations and vision fallback for unknown places.

polt-skill connects agents to the POLT collaborative project and task platform.

Crypto Wallets & Payments for AI Agents documents crypto wallets, token transfers, and payment workflows for agents.

Provide real-time monitoring and workspace views for OpenClaw agents.

Manage authorized project memory without changing registry truth.

Save, switch, or compare Codex accounts locally, with optional sync to OpenClaw.

AI 技能自动发现、评估、集成、验证、宣传闭环系统 — 跨生态技能市场引擎

Create, list or cancel reminders and cron jobs scoped to the originating channel.

定位：这是你之前接触的零散医药AI技能的「总调度中枢」——不是单一的内容生成工具，而是站在「医药营销内容运营总监」视角，把之前聊过的DA生成、PPT制作、话术设计、KV创作等所有单点技能，串成一条「内容生产→能力转化→合规闭环→持续进化」的全链路智能平台。相当于交响乐团的指挥，让各个单点技能协同工作，解决过去「工具零散、数据不通、产训两张皮、合规各管各」的行业痛点。 核心价值：解决医药内容生产的4大顽疾

ElevenLabs Agents: Manages conversational AI agents through ElevenLabs, including listing, creation, configuration, and deployment.

gws-calendar is a guide to managing Google Calendar calendars and events through the gws command-line tool.

Register an AI agent and chat with other randomly matched agents.

Back up OpenClaw identity, configuration, and memory files to the Everclaw remote service, and restore files when they are missing locally.

A pluggable self-improvement framework for AI agents. Automatically learns from mistakes, corrections, and feedback to continuously improve execution quality...

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express...

An integration for invoking Hermes Agent through its CLI.

Add lightweight causal analysis to agent actions by modeling interventions, counterfactuals, and observed outcomes.

agent-team-kit provides role, work-queue and feedback templates for multi-agent teams.

Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.

Control the user's real logged-in browser through the local Kimi WebBridge daemon with safe cross-platform helpers, bounded snapshots, conditioned waits, and recovery diagnostics. When both kimi-webbridge and kimi-webbridge-pro are installed, prefer this Pro skill for helper-driven browser workflows. Use when the task requires a real browser session, login state, existing tabs, screenshots, form filling, confirmed file uploads, PDF saving, or network diagnosis. Prefer a dedicated API, MCP tool, or site-specific skill when one is explicitly available and sufficient. Do not use for pure web search, factual lookup, or tasks that do not need browser state.

Skill Manager manages the lifecycle of installed skills by suggesting relevant skills from task context, tracking installations, checking updates, and guiding cleanup.

Describes a commercial framework for agent orchestration, memory, fallback and self-improvement workflows.

Ralph Mode Autonomous Development Loop Skill

Autonomous experiment loop for AI agents. Use when the user wants to run systematic experiments — optimizing hyperparameters, searching for better configurat...

Top-level macOS computer-use skill with a bundled standalone runtime that bootstraps itself without any local Claude installation, private native modules, or...

X Spaces, but for AI Agents. Live voice rooms where AI agents host conversations.

A skill that lets an agent place outbound calls with an AI voice agent.

每天自动读取 GET笔记 中昨日的录音记录，生成「柳比歇夫日记+发芽报告+复盘总结」三合一日报，推送到飞书/微信等 IM 会话。

Free, local, read-only security self-audit for your own OpenClaw agent. Reads your OpenClaw config, bootstrap files, logs, agent session logs, and installed skills, plus a bounded host-security scan. It changes no OpenClaw config, skill or bootstrap file; its own history and state stay under ~/.clawseccheck (--no-history skips it, --purge removes it). Grades your setup A-F when all five check layers ran, naming what's missing otherwise. --monitor records a local baseline so every later run alerts on what changed - a new MCP server, an edited skill, config drift, a finding that appeared or cleared. No API key, no network calls; the only external command it runs is your own read-only openclaw security audit (skip with --no-native). --save and --html write only to the path you give. Use it when you want to check or audit your OpenClaw agent's security, find prompt-injection or misconfiguration risks, see your A-F security score, watch your OpenClaw setup for changes, or ask what changed since the last check.

达尔文.skill 2.0 — 自主Skill优化系统：评估→改进→测试→保留或回滚。与女娲.skill配合使用：女娲造人（创建Skill），达尔文进化（优化Skill）。集成微软SkillLens 9维评分+SkillOpt验证机制

flow is an intelligent skill orchestrator for Clawdbot/MCP that compiles natural language requests into secure, reusable FLOW workflows.

Smooth Browser provides a CLI for agents to navigate websites, authenticate, extract data, and perform web workflows.

Automate web browsing with navigation, form filling, clicking, screenshots, data extraction, and testing using Chrome via OpenClaw browser tool.