full-stack-skills/agent-skills

skill-sop-creator

Guide for creating Standard Operating Procedures and SOPs and composite workflows by combining atomic Agent Skills.

First seen Jun 6, 2026

Installation

$ npx skills add full-stack-skills/agent-skills --skill skill-sop-creator

Summary

  • Guide for creating Standard Operating Procedures and SOPs and composite workflows by combining atomic Agent Skills.
  • Use when users want to design, structure, or implement multi-step workflows that orchestrate multiple skills for complex business goals.
  • Transforms loose tool collections into reliable, repeatable business processes.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from full-stack-skills/agent-skills · top by installs.

npx skills add full-stack-skills/agent-skills

Browse all from full-stack-skills/agent-skills

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

Stars 2
License LICENSE
Default branch main
Status Active

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 3,567 B
  • docs SUMMARY.md 357 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 35 installs

SKILL.md

skill-sop-creator

Guide for creating Standard Operating Procedures (SOPs) and composite workflows by combining atomic Agent Skills. Use this skill when users want to design, structure, or implement multi-step workflows (SOPs) that orchestrate multiple skills to achieve complex business goals. This skill transforms loose collections of tools into reliable, repeatable business processes.

Parameters

Name Type Required Description
action string true The action to perform. One of searchskills, draftsop, validate_sop.
context string false Context for the action (e.g. business goal for drafting, search query for searching).

Actions

1. Search Atomic Skills

Finds available atomic skills to include in your SOP.

  • Command: skill-sop-creator(action="search_skills", context="pdf processing")
  • Output: List of relevant atomic skills (e.g., pdf-split, pdf-merge, ocr-extract) that can be steps in your SOP.

2. Draft SOP Structure

Generates a draft SKILL.md for a composite skill (SOP) based on a business goal.

  • Command: skill-sop-creator(action="draft_sop", context="Daily Competitor Analysis Report")
  • Output: A structured markdown template defining the workflow steps, required atomic skills, and decision logic.

3. Validate SOP

Checks if a proposed SOP structure is valid and if referenced skills exist.

  • Command: skill-sop-creator(action="validatesop", context="<sopcontent>")
  • Behavior: Verifies that steps are logical and dependencies are met.

SOP Design Principles

  1. Atomic Foundation: SOPs should rely on small, single-purpose skills (e.g., pencil-draw-rect not draw-entire-ui).
  2. Deterministic Flow: Define clear steps: Step 1 -> Step 2 -> Decision -> Step 3.
  3. Human-in-the-loop: Include checkpoints where the Agent should ask for user confirmation.

Example SOP Structure

# Daily Competitor Analysis SOP

## 1. Information Gathering
- **Skill**: `web-search`
- **Action**: Search for "competitor name" + "news"
- **Output**: List of URLs

## 2. Content Extraction
- **Skill**: `web-scraper`
- **Input**: URLs from Step 1
- **Output**: Raw text

## 3. Analysis
- **Skill**: `llm-summarize`
- **Input**: Raw text
- **Prompt**: "Identify key pricing changes and feature launches."

## 4. Reporting
- **Skill**: `pdf-generator`
- **Input**: Summary
- **Output**: PDF Report

Notes

  • An SOP is essentially a "Composite Skill" that acts as a conductor for other "Atomic Skills".
  • This skill helps you design that conductor.

国内适配

  • 支持中文文档和中文注释
  • 示例代码兼容国内开发环境
  • 提供中文 FAQ 和常见问题解答

能力边界

✅ 适用场景

  • 当你需要使用此技能对应的技术栈时
  • 当项目需要遵循最佳实践时
  • 当需要快速上手或深入理解核心概念时

⚠️ 需要注意

  • 复杂业务逻辑需要结合具体场景调整
  • 性能优化需要根据实际数据量评估

❌ 不适用场景

  • 不相关的技术栈或框架
  • 需要完全自定义的特殊场景