The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 354 files from 1 739 authors, of which 61 713 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
Pragmatic clean code standards for Delphi — concise, direct, no over-engineering
Threading patterns in Delphi — TThread, TTask, TParallel, Synchronize, Queue, thread-safety, Producer-Consumer, pools, cancellation and debugging
Good memory management practices, memory leak prevention and exception handling in Delphi
Delphi code review checklist — quality, security, performance, SOLID, memory
SOLID implementation patterns for Delphi projects — Repository, Service, Factory, Strategy with constructor injection and interfaces
Implementation of the 23 GoF (Gang of Four) patterns in Object Pascal / Delphi with interfaces, TInterfacedObject and SOLID principles. Covers Creational, Structural and Behavioral patterns.
Standards for using DevExpress (DEXT) components in Delphi VCL applications
Unit testing patterns with DUnitX for Delphi — fixtures, mocking, integration tests
Architectural patterns, Entity ORM, Minimal APIs and dependency injection for projects created with Dext Framework (cesarliws/dext).
Patterns for development with DelphiMVCFramework (DMVC) — controllers, Active Record, JWT, Swagger
Patterns for developing REST APIs with Horse framework in Delphi
Guides and standards for using the Intraweb stateful web framework in Delphi projects.
Development patterns with Firebird database via FireDAC — connection, PSQL, generators, transactions, migrations
Development patterns with MySQL/MariaDB via FireDAC — connection, stored procedures, AUTO_INCREMENT, JSON, triggers, replication, migrations
Development patterns with PostgreSQL via FireDAC — connection, PL/pgSQL, sequences, JSONB, UPSERT, full-text search, migrations
Refactoring techniques for Object Pascal: Extract Method, Extract Class, Guard Clauses, Replace Magic Numbers, Replace Conditional with Polymorphism, Introduce Parameter Object, Remove With, Extract Interface. Always focusing on maintaining behavior and improving readability.
Guidelines on how the AI should act and code when the user requests TDD, unit tests, DUnitX or fakes/mocks using Interfaces in Delphi.
Threading patterns in Delphi — TThread, TTask, TParallel, Synchronize, Queue, thread-safety, Producer-Consumer, pools, cancellation and debugging
Use when user requests diagrams, flowcharts, sequence diagrams, class diagrams, component diagrams, ER diagrams, architecture charts, or visualizations — including generating a diagram from existing source code, or rendering/extracting PlantUML embedded in a Markdown file to images (e.g. preparing docs for Confluence/Notion). Also use proactively when explaining systems with 3+ components, APIs, data flows, or class hierarchies. Generates .puml files and exports to PNG/SVG via Kroki API (no local install required).
根据用户描述生成高质量绘图 prompt,并按通用、roadmap、schematic 模式调用 gpt-image-2 或 Nano Banana/Gemini 图片模型 API;gpt-image-2 默认使用低画质、原生尺寸和 JPEG,第 2 轮起基于上一轮图片做保真微调。
当用户明确要求"测试代码"、"运行代码审查"或"进行代码自检"时使用。通过多轮 A 轮批判性代码审查 + B 轮代码质量原则检查,系统化发现、记录、修复程序代码中的问题,并将计划/过程/结果统一沉淀到目标代码根目录的 `.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/auto-test-code/{yyyy-mm-dd-hh-mm}/output/tests/` 隔离工作区。⚠️ 不适用:用户只是想优化功能(应直接修改)、只是询问代码问题(应直接回答)、没有明确"测试代码"意图。
当用户明确要求"使用 brainstorming"或"使用 awesome-code"时使用。⚠️ 不适用:用户只是想优化/改进某个功能(应直接修改)、只是询问技能问题(应直接回答)、没有明确使用 brainstorming/awesome-code 的一般性开发。
当用户明确要求"测试技能"、"运行 auto-test"或"进行批判性测试"时使用。通过多轮 A 轮批判性测试 + B 轮质量原则检查,系统化发现、记录、修复问题,并沉淀可追溯的 `.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/auto-test-skill/output/plans/` 与 `.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/auto-test-skill/output/tests/` 文档。⚠️ 不适用:用户只是想优化功能(应直接修改)、只是询问技能问题(应直接回答)、没有明确"测试"意图。
Use when completing tasks, implementing major features, or before merging to verify work meets requirements - reviews implementation against plan or requirements with severity分级(Critical/Important/Minor). NO MERGE WITHOUT CODE REVIEW FIRST.
当用户明确要求"使用 awesome-code / 多代理协作 / 并行协调开发"时使用。通过脚本收集可用 Agent 摘要、配置约束与 `dispatch_gate`,再由 AI 自主判断 single-pass / focused-agent / parallel / sequential 策略并选择子代理;当配置中的 required route agent 缺失时必须阻塞继续执行。⚠️ 不适用:用户仅需单一角色的简单修改或咨询、用户未明确表达多代理协作意图、用户只是了解技能概念。
当用户明确要求"测试项目"、"运行 auto-test-project"或"进行项目级测试"时使用。对完整项目进行多轮 A 轮批判性测试 + B 轮质量检查,系统化发现、记录、修复问题。⚠️ 不适用:用户只是想优化功能(应直接修改)、只是询问项目问题(应直接回答)、没有明确"测试"意图。
后端开发专家。精通 Node.js/Python/Go/Rust 等后端技术栈,专注于 API 设计、数据库优化、认证授权、微服务架构和性能调优。用于后端服务开发、API 设计和系统架构。
上下文优化专家。专注于长对话中的上下文管理、token 效率和性能优化。解决 lost-in-middle、context poisoning 等问题,提升 AI 代理在复杂任务中的表现。
文档专家。专注于技术文档编写、API 文档生成、README 优化和文档维护。提供清晰的文档结构、规范的格式和用户友好的内容。
DevOps 与运维专家。精通 CI/CD、容器化、编排、基础设施即代码、监控告警和自动化部署。用于构建高效、可靠的软件交付流水线和运维系统。
Git 工作流专家。规范化版本控制,确保提交历史清晰可追溯。支持 Conventional Commits 规范、Pull Request 最佳实践、分支管理策略和自动化工作流。
Use when executing implementation plans with independent tasks - dispatches fresh subagent for each task with code review between tasks, enabling fast iteration with quality gates. Supports orchestrator, peer-to-peer, and pipeline coordination modes.
前端开发专家。用于前端应用、页面、组件、现有 UI 增量改进、已有页面视觉优化、UI/UX 改造、视觉重设计、响应式布局、表单和输入框视觉一致性、设计系统落地、React/Vue/Next.js/Nuxt、Tailwind、shadcn/ui、Radix、MUI、AntD、仪表盘、SaaS、Landing Page、移动端适配、前端性能优化、浏览器截图检查和去除 AI 味 UI。触发后应把“做好看”转化为可执行设计约束,优先复用项目现有组件与视觉语言,小步实现并尽量通过浏览器或 Playwright 做视觉反馈闭环;当用户要改进当前页面或已有界面时,默认先诊断再做局部优化,而不是从 0 重做。
当用户明确要求"优化镜像源"、"配置国内镜像"、"加速部署"、"切换镜像源"或项目涉及 Docker/Python/Node.js/Go/Java/Ruby/Rust 等需要包管理器的场景时使用。自动检测项目使用的包管理器,生成适配的国内镜像源配置,确保在国内网络环境下快速部署。⚠️ 不适用:用户只是想了解镜像源概念(应直接说明)、项目已完成部署且无需优化(应直接告知无需操作)、用户明确要求使用官方源(应尊重用户选择)。
安全专家。专注于应用安全、威胁建模、安全合规和数据保护。提供安全审查、漏洞扫描、安全配置和合规检查。用于构建安全可靠的应用系统。
Use when asked to create, review, or revise a multi-step implementation, bug-fix, or change plan before editing code, including existing plans that may use a legacy task-by-task template. Produce a problem-first plan that non-technical readers and implementers can both understand.
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes - four-phase framework (root cause investigation, pattern analysis, hypothesis testing, implementation). NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST.
Use when implementing any feature or bugfix, before writing implementation code - write the test first, watch it fail, write minimal code to pass; ensures tests actually verify behavior by requiring failure first. Enforces RED-GREEN-REFACTOR cycle with iron-law compliance.
当用户明确要求"优化 prompt"、"改进提示词"、"润色指令"或"将简陋 prompt 转换为最佳实践版本"时使用。基于 OpenAI 和 Anthropic 官方最佳实践,对用户提供的简陋 prompt 进行结构化优化,输出符合社区标准的高质量版本。
当 Bensz 系列 skills 在真实用户环境中因 skill 设计缺陷而出现 bug,或用户明确说“我想 report bensz skills bugs”“帮我公开上报 bensz skills 的 bug”时使用。该 skill 负责把 bug 规范化记录到 `~/.bensz-skills/bugs/`,并在用户明确要求公开报告时通过本地 `gh` 轻量上传到 `huangwb8/bensz-bugs`,全程严禁修改用户本地 Claude Code/Codex 中已安装 skills 的源代码。
当用户明确要求“压缩/瘦身/精简某个 Agent Skill 的 Markdown 文档”“在不改变功能前提下降低 skill 上下文开销”时使用。先理解目标 skill 的真实能力与安全边界,再在忽略 `tests/`、`plans/` 以及目标 skill 的 `README.md`、`CHANGELOG.md` 的前提下,压缩 `SKILL.md`、`references/*.md` 等工作型 Markdown,并把中间产物隔离到 `.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/compact-bensz-skills/`。⚠️ 不适用:用户主要想新增功能、修复脚本逻辑、批量改代码、或只想压缩非 skill 文档。
当用户明确要求"提交 Git 改动"、"生成 commit 信息"或"创建 git commit"时使用。仅用 Git 分析改动并自动生成 conventional commit 信息(可选 emoji);必要时建议拆分提交,默认运行本地 Git 钩子(可 --no-verify 跳过),提交后默认自动 push(可 --no-push 跳过)。
当用户明确要求“review 某个 GitHub PR”“评估某个 pull request 是否值得 merge”“帮我判断这个 PR 怎么处理”时使用。基于用户提供的 GitHub 仓库地址、PR 编号/链接和补充说明,进行只读、证据驱动的 PR 审查:理解 PR 解决的问题、评估方案优劣与局限、默认优先使用内置“好 PR”标准并在必要时联网补充、识别恶意或高风险改动,并输出是否建议 merge 的 Markdown 决策报告。⚠️ 不适用:用户要你直接修改 PR 代码、直接 merge PR、或在本地执行 PR 分支中的不可信代码。
当用户明确要求"初始化项目"、"创建项目指令文件"或"生成 AGENTS.md"时使用。完全自动化:自动检测操作系统默认语言,分析项目目录结构(支持 Python/Web/Rust/Go/Java/数据科学/文档项目等),推断项目类型和用途,一键生成规范的项目指令文档。生成结果包括:AGENTS.md(跨平台通用项目指令,Single Source of Truth)、CLAUDE.md(Claude Code 特定适配,通过 @./AGENTS.md 引用)、README.md(项目介绍与使用方法)、CHANGELOG.md(项目变更记录)、.gitignore(Git 忽略规则,安全优先),并在完整初始化时自动补齐 `docs/` 与 `docs/plans/`。
当用户明确要求"发布项目到 GitHub"、"创建 GitHub Release"或"生成 Release Notes"时使用。智能分析 tag 间历史变化,生成专业且吸引人的 Release Notes,自动创建 GitHub Release。支持首次发布、常规版本、预发布版本(alpha/beta/rc),自动识别 prerelease 标记。
当用户明确要求"并行执行同一条 Vibe Coding 指令 / 多个独立 agent 或 subagent 同时审查、想方案、优化、对比多条路线 / 多线程独立尝试"时使用。默认使用智能模式:由宿主原生 subagent 独立分析并由主 agent 汇总;智能模式和代码模式必须使用同一套 `.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/parallel-vibe/{yyyy-mm-dd-hh-mm}/` 运行目录、`@main/plan.json`、thread `workspace/`、`RESULT.md` 与 `runner.log` 契约,区别只在底层执行机制;当用户要求脚本 runner、plan-file、resume、跨 CLI runner、退出码或无可用 subagent 时,切换到代码模式并调用 `scripts/parallel_vibe.py`。⚠️ 不适用:普通 shell 并发、单元测试并发、下载任务、要求强安全隔离或处理高度敏感数据。
当需要把本仓库 pipelines/skills 下的所有 skills 安装到系统级(默认同时安装到 Codex: ~/.codex/skills 和 Claude Code: ~/.claude/skills),以便在任意项目/对话中可被发现与调用时使用。使用 MD5 哈希进行版本控制,仅安装有更新的 skills;支持 --skill 指定单个或少量技能安装/更新、强制覆盖安装、指定单一目标安装和远程安装模式(--remote --check/--auto)。
当用户明确要求"生成技能 README"、"编写用户指南"或"更新技能文档"时使用。为 Agent Skills 编写符合最佳实践的 README.md 用户使用指南。自动分析技能结构(SKILL.md、config.yaml、scripts/),按模板生成小白友好的文档。
| 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. cross-partition queries, SDK singleton, CosmosClient, container modeling, change feed, bulk operations, vector search, full-text search, hierarchical partition keys, global distribution, autoscale throughput, indexing policy. Azure SQL, Cosmos DB for PostgreSQL (vCore), Cosmos DB for MongoDB vCore, Azure DocumentDB, general SQL databases, Redis, Elasticsearch.
Detect Frappe project type, installed apps, version, and tooling. Use as the first step when working on any Frappe/ERPNext codebase to understand the project structure before making changes.
Production-grade architectural patterns for building enterprise Frappe apps like CRM, Helpdesk, and HRMS. Use when designing complex multi-entity systems with workflows, SLAs, and integrations.
Build REST and RPC APIs in Frappe including whitelisted methods, authentication, and permission handling. Use when creating custom endpoints, integrating with external systems, or exposing business logic via API.
Customize Frappe Desk UI with form scripts, list view scripts, report scripts, dialogs, and client-side JavaScript APIs. Use when building interactive Desk experiences, adding custom buttons, or scripting form behavior.
Create and modify Frappe DocTypes including schema design, controllers, child tables, and customization. Use when building data models, adding fields, or implementing document lifecycle logic.
Scaffold and architect custom Frappe apps including app structure, hooks, background jobs, service layers, and production hardening. Use when creating new apps, setting up app architecture, or implementing cross-cutting patterns like caching, logging, and error handling.
Use Frappe Manager (FM) for Docker-based development and testing environments. Use when setting up local dev, running isolated tests, or managing agent-driven Frappe development workflows.
Build modern Vue 3 frontend apps using Frappe UI with components, data fetching, and portal pages. Use when creating custom frontends, SPAs, or portal interfaces for Frappe applications.
Create reports in Frappe including Report Builder, Query Reports (SQL), and Script Reports (Python + JS). Use when building data analysis views, dashboards, or custom reporting features.
Route to the appropriate Frappe skill based on task type. Use as the entry point when working on Frappe projects to determine which specialized skill to apply.
Write and run tests for Frappe apps including unit tests, integration tests, and UI tests. Use when adding test coverage, debugging test failures, or setting up CI for Frappe projects.
Answers built from the skills we actually parsed.