Automation
Task automation and scripting
DOCX creation, editing, and analysis
by yousufjoyian
Claude agent skills and workspace configuration
manus
by hqman
Delegate complex autonomous tasks to Manus AI - an AI agent for deep research, web browsing, code execution, report generation, and multi-step workflows.
viem-sweep
by melonask
Comprehensive guide and reference for implementing advanced sweeping and transaction strategies (Legacy, Factory, Permit, Auth, EIP-7702) using viem v2. Use this skill when you need to implement token sweeping, wallet factories, gasless transfers, or account delegation.
codex-opencode-deepseek-workflow
by ysj98
Explicit-only personal Codex skill for any Git project when the user asks for OpenCode/DeepSeek execution or names $codex-opencode-deepseek-workflow. Codex creates a lightweight AI development task order, invokes OpenCode with the configured worker model in the current working directory, and stops for human code review, runtime validation, and Git finalization.
root-cause
by Nomik94
Root Cause Analyst 에이전트 스폰. Use when: /root-cause, 버그 원인 찾아줘, 왜 안 돼, 이상 현상, 간헐적 에러, 재현 어려운 문제, 원인 분석, 디버깅 도와줘. NOT for: 단순 에러 메시지 해석, 기본적인 typo 수정.
decompose-task
by fancy-bread
Decompose Task
project-workflow
by zhucl1006
文档驱动的项目开发工作流。智能判断任务类型，执行已有计划或使用 project-planning 生成新计划。 使用场景： - 已有完整项目文档，需要执行开发任务 - 用户指定执行某个计划（如"执行 plan 001"） - 用户请求开发新功能（使用 project-planning 生成计划） 前置条件： - 项目应该已有 docs/ 文档结构（PRD、SAD 等） - 如果是新项目，建议先使用 project-docs-setup skill 创建完整文档 - 如果需要制定计划，建议使用 project-planning skill 与其他 skill 的关系： - project-docs-setup：创建项目文档结构 - project-planning：制定开发计划（需求澄清 + 设计讨论 + 计划编写） - project-workflow：执行开发计划 + 更新文档 - 建议流程：project-docs-setup → project-planning → project-workflow 触发方式： - "执行 plan 001" / "继续 001-user-authentication" - "开始开发" / "执行计划" / "继续上次的任务"
feature-planner
by Nomik94
Use when: 기능 구현 계획, 구현 전 설계, 스코프 정의, 작업 분해, 3개 이상 파일 변경, 기능 추가, feature planning, 구현 어떻게 할까, 큰 작업, 복잡한 기능, 어디부터 시작, 작업 순서. NOT for: 단순 버그 수정, 1-2 파일 변경, 오타 수정.
amplitude-automation
by ncdevshiv
"Automate Amplitude tasks via Rube MCP (Composio): events, user activity, cohorts, user identification. Always search tools first for current schemas."
proxy-manager
by DrowningDuck0
"AI-assisted proxy management for WSL environments. Manages clash-meta proxy lifecycle, automatic speed testing for best node selection, and task-level proxy isolation. Use when a project requires external network access (pip install, git clone, curl downloads, huggingface). Covers the full workflow from proxy need detection to user approval to task execution."
cx-plan
by m19803261706
CX 工作流 — 任务规划与契约下沉。当用户提到"规划任务"、"制定计划"、 "plan"、"拆分任务"时触发。读取 Design Doc（或 PRD for S 规模） 生成任务分解，创建子任务文件 task-{n}.md，每个任务包含设计文档中的 API 契约片段。保存到本地 .claude/cx/features/{dev_id}-{feature}/tasks/。
subtask-orchestration
by ApocalypseYun
Orchestrate parallel subtasks across multiple repositories using vibe-kanban MCP. Use this skill when: (1) Work needs to be distributed across multiple repos, (2) Parallel execution of independent tasks is needed, (3) You need to create subtasks and collect their results, (4) Multi-agent coordination via kanban task descriptions. Triggers: "subtask", "multi-repo", "parallel tasks", "distribute work", "orchestrate".
parallel-task
by yuanxiao0115
Execute plan files by launching multiple parallel subagents to complete tasks simultaneously. Triggers on explicit "/parallel-task" commands.
insights-archive
by jonathanprozzi
Archive Claude Code /insights reports for historical tracking. Supports archiving, listing, comparing, and opening reports.
browser-ops
by Garenaof2462
Browser automation and web extraction platform for OpenClaw. Use when the user wants to automate websites, click/type/fill forms, capture screenshots, crawl list/detail pages, paginate, extract structured data, observe page/network behavior, or build reusable browser-driven workflows. Prefer for browser-based operations that need stronger execution, recovery, logging, checkpoints, site profiles, or mixed automation + crawling flows. Do not use to bypass captchas, evade access controls, or break platform security mechanisms; instead use human-in-the-loop checkpoints for login, MFA, captcha, or approval walls.
ataski
by Hebilicious
Manage project work with the Ataski filesystem board. Use when users ask to create, track, claim, move, or complete tasks; coordinate sub-agents with git worktrees; enforce task dependencies with blockedBy; or maintain ataski/tasks.md.
executing-plans
by aleister1102
Use when you have a written implementation plan to execute in a separate session with review checkpoints
phase-plan
by biruk741
Create a detailed, phased execution plan with tasks assigned to specific team members by role. Handles dependency mapping between tasks, identifies parallelizable work, sets gate points where work must sync, and produces per-person task boards. Use when the team is ready to start building and needs to know who does what and in what order. Also use for re-planning mid-hackathon if priorities shift.
skill-value-assessor
by qingchunwuhui
Evaluate whether a task is worth converting into an AI Skill using a three-dimensional assessment model (Cognitive Friction, Process Structure, Reuse Value). Use when the user wants to decide if a workflow should be automated as a Skill, asks questions like "Should I make this a Skill?", "Is this worth automating?", or explicitly invokes with /assess-skill or /skill-eval. Provides guided questioning, automatic scoring, and concrete recommendations.
gsDesignNB
by keaven
Design, simulate, and adapt clinical trials with negative binomial recurrent event endpoints using the gsDesignNB R package. Use this skill when the task involves: NB sample size or power, event gaps and Jensen correction, calendar-time group sequential design, blinded or unblinded information estimation, sample size re-estimation (SSR), recurrent-event simulation (constant or seasonal rates), interim data cuts, completer-based analyses, or non-inferiority/super-superiority designs for recurrent events.
devdocs-dev-workflow
by ab300819
Execute development tasks with skeleton-first approach and layered TDD. Supports single task, batch execution (by range, feature, user story), dependency resolution, and breakpoint resume. Includes optional adversarial verification and --headless unattended mode (无人值守). Triggers on "execute task", "start T-XX", "batch", "resume", "开发任务", "执行任务", "批量开发", "继续开发", "--review", "--headless", "无人值守".
norske-turnusregler
by GauteR
Sette opp og vurdere turnus, arbeidsplan og vaktliste ut fra norsk lov for stat, kommune og privat; lese Excel og generere forslag til fordeling; arbeidstidsgrenser og fordeling deltid/heltid etter stillingsprosent. Brukes når bruker spør om turnus, vaktliste, arbeidsplan, skift, Excel-turnus, fordeling, deltid, stillingsprosent, arbeidstid, AML, Arbeidstilsynet, stat, kommune eller privat.
Video Slides Production - 视频幻灯片生产技能
by zhangmaosen
让内容自己表达
competing-hypotheses
by shhac
Debug problems by investigating multiple hypotheses in parallel. Use when you have a bug, unexpected behaviour, or mystery where the root cause is unclear. Spawns parallel investigator agents each pursuing a different theory, then compares evidence to identify the most likely cause and fix.