agent

在当前 Printing Press 下从头重新生成已发布的 CLI，将先前的研究和先前的新特性带入 novel-features 子代理的重印协调流程中，而不是直接丢弃。如果本地不存在该 CLI，将从公共库中拉取；根据存在时间建议复用还是重做先前的研究；然后携带正确的上下文移交给 /printing-press。当机器升级对已发布 CLI 的提升大于手动优化时使用。触发短语："reprint <api>"、"regenerate <api>"、"redo the <api> CLI"、"rebuild <api> from scratch"、"this CLI would benefit from a reprint"。

Create skills by learning from high-quality GitHub projects

Master core Java programming - syntax, OOP, collections, streams, and exception handling

Build AI agents that interact with computers like humans do - viewing screens, moving cursors, clicking buttons, and typing text. Covers Anthropic's Computer…

Guide agents through structured research including planning, multi-query execution, source analysis, and synthesis. Use for comprehensive topic research, deep…

Comprehensive guide to sub-agents in Claude Code: built-in agents (Explore, Plan, general-purpose), custom agent creation, configuration, and delegation patterns. Use when: creating custom sub-agents, delegating bulk operations, parallel research, understanding built-in agents, or configuring agent tools/models.

将数据导入 AWS 数据湖，支持从 S3 文件、本地上传、JDBC 数据库（Oracle、SQL Server、PostgreSQL、MySQL、RDS、Aurora）、Amazon Redshift、Snowflake、BigQuery、DynamoDB 或现有 Glue Catalog 表（迁移）。默认目标为 S3 Tables；未采用 S3 Tables 时支持通用存储桶上的标准 Iceberg。处理一次性加载、周期性管道、迁移。触发场景：导入数据、加载数据、摄取、同步数据库、迁移表、将数据迁移至 AWS、设置管道、ETL、从 Snowflake 拉取、将 BigQuery 查询结果写入 S3、导出 DynamoDB、CTAS、转换为 Iceberg。请勿用于设置或排查 Glue 连接（使用 connecting-to-data-source）、创建空表（使用 creating-data-lake-table）、运行查询（使用 querying-data-lake）、模糊搜索表名（使用 finding-data-lake-assets）、Catalog 审计（使用 exploring-data-catalog），或 Salesforce、ServiceNow、SAP、MongoDB、Kafka 等 SaaS 平台。

内部子技能：对打印的 CLI 采样命令输出进行智能审查，以排查基于规则的检查无法编码的合理性问题（子串匹配相关性、格式错误、静默源丢弃、排序失败）。在诊断循环期间，通过 Skill 工具由主 printing-press SKILL.md（阶段 4.85）和 printing-press-polish SKILL.md 调用。不供用户直接调用——其实用的封装入口为 /printing-press 和 /printing-press-polish。

Wire Alchemy into application code without an API key, using the x402 or MPP gateway with wallet-based auth (SIWE/SIWS) and per-request payments (USDC via…

Pay people in SOL or USDC, buy and sell tokens, check prices, discover trending and new tokens, create and manage Solana wallets, stake SOL, earn yield through lending and managed vaults, borrow against collateral, set up DCA (dollar-cost averaging) and limit orders, provide liquidity across multiple DEXes, trade prediction markets, pay for APIs via x402, set up security permissions and spending limits, and track portfolio performance — all from the command line. No API keys, no private key env vars. Use when the user wants to send crypto, swap or trade tokens, browse what's trending, check balances, earn yield, borrow, set up recurring buys or limit orders, provide liquidity, bet on predictions, pay for web resources, or see how their holdings are doing.

Designs graphics for thumbnails, social media, banners, and presentations. Applies design principles (CRAP, Gestalt, visual hierarchy) with research-backed…

Azure Cosmos DB performance optimization and best practices guidelines for NoSQL, partitioning, queries, and SDK usage. Use when writing, reviewing, or refactoring code that interacts with Azure Cosmos DB, designing data models, optimizing queries, or implementing high-performance database operations.

Make websites accessible for AI agents. Navigate, click, type, extract, wait — using Chrome with existing login sessions. No LLM API key needed.

Generate an explorable HTML report of Claude Code session usage (tokens, cache, subagents, skills, expensive prompts) from ~/.claude/projects transcripts.

从工作流描述创建跨平台智能体技能。当用户要求创建智能体、自动化重复工作流、创建自定义技能或需要高级智能体创建时激活。触发短语包括：create agent for、automate workflow、create skill for、every day I have to、daily I need to、turn process into agent、need to automate、create a cross-platform skill、validate this skill、export this skill、migrate this skill。支持单技能、多智能体套件、转录处理、基于模板创建、交互式配置、跨平台导出和规格验证。

主应用构建编排器。通过自然语言请求创建全栈应用。确定项目类型，选择技术栈，…

重构臃肿的 AGENTS.md、CLAUDE.md 或类似的 agent 指令文件，遵循渐进式披露原则。将单体文件拆分为结构化的…

Remove AI-generated code slop from the current branch. Use after writing code to clean up unnecessary comments, defensive checks, and inconsistent style.

Context optimization extends the effective capacity of limited context windows through strategic compression, masking, caching, and partitioning. The goal is…

在合并分支或完成任务后使用，将经验总结整合到记忆层，关闭问题并验证阶段关卡。