OpenClaw AI 技能

检查 how much Claude Max / Claude Pro subscription quota you have LEFT — not how much you spent. Most usage 工具s 追踪 API billing or 令牌 costs. This one ans

安全 MCP bridge enabling Claude on your phone to browse and edit local repos with real-time, human-应用roved file 访问 and 审计 记录ging.

Build autonomous AI 代理s with Claude 代理 SDK. Structured 输出s guarantee JSON 模式 验证, with 插件s 系统 and hooks for event-driven 工作流s. 预防s 14 documented errors

Evaluate and 验证 Claude Code rules in .claude/rules/ directories. Use for frontmatter, glob patterns, and 质量 审计s

Helps 创建, 更新, and 审计 OpenClaw 技能s to ensure high 质量 following Anthropic's best practices; triggered by phrases like "创建 a new 技能" o...

A universal MCP-compatible 工具/技能 that empowers ANY AI 代理 to natively call the local Anthropic Claude Code 命令行工具. Ideal for code refactoring, large 扩展

Claude Code 终极使用指南 - 涵盖上下文工程、规范驱动开发、Hooks 回调、Sub 代理s、输出 Styles、SuperClaude 框架等高级技巧。当用户需要：1) 提升 Claude Code 编程效率，2) 省 令牌 技巧，3) Spec-driven 开发，4) 多 代理 协

停止 Claude Code from repeating mistakes — enforce 防护rAIls, preserve 上下文, mAIntAIn consistency across 会话s

Analyze Claude Code 令牌 usage. Shows where 令牌s went, which projects cost most, and how to reduce waste. Use when user asks about 令牌 usage, costs, or bu

进化版统一记忆系统，融合Claude Code的5层记忆架构+令牌预算管理+智能成本追踪。当用户要求记忆跨会话信息、智能管理令牌预算、优化AI调用成本、构建个人知识库、跨项目学习进化时使用。

Reduce LLM API 令牌 consumption by 20-35% through pre-发送 estimation, memory 提取ion, and 上下文 压缩ion.

安装 and 配置 a custom Claude Code 状态 line showing real-time 令牌 usage, 上下文 window percentage, git branch, and color-coded 警告s. Use when the user asks to "

Switch OpenClaw heartbeat 模型 to local Ollama and 部署 a 防护 to 预防 un授权d changes, avoiding pAId cloud 令牌 usage.

令牌-efficient 上下文 格式化 using LLMCOM specification - reduces 令牌 usage by 70-80% through compact object notation.

追踪, analyze, and 优化 LLM API spending across 提供者s (OpenAI, Anthropic, Google, DeepSeek, etc.). Use when the user asks about API costs, 令牌 usage, billin

监控s and enforces real-time LLM API spending with per-user quotas, alerts, dAIly 报告s, and bud获取 limits to 预防 cost over运行s.

技能 自动评估和改进管线。9 维结构评分（含 LLM-as-Judge）、4 角色加权、 类别修正系数（工具/knowledge/orchestration/rule）、Pareto front 回归保护 （security 2%/efficiency 10%/其他 5%）、追踪-aware 失败.

Replace text in PDF files while preserving visual fidelity. Handles custom font encodings, embedded sub设置s, 加密ed PDFs, image-based 扫描s (OCR), variable

AEO analyst orchestration — coordinates canonry sweeps and aeo-审计 analysis with persistent memory and proactive regression 响应.