ai-workflow
AGENTS.md283 starsChanged 10 months agoMITView on GitHub
This is the master repository for curated Claude Code workflow collections. IMPORTANT: All AGENTS.md files in this project MUST be written in English only. This ensures consistent AI behavior across different language contexts. Creates complete workflow directories with: - Multi-AI tool support (.claude, .codex, .cursor, .opencode, .agents, .kilocode, .roo, .goose, .gemini, .agent, .github, skills, .factory, .windsurf) - Curated skills downloaded from GitHub - README.md (user documentation) - AGENTS.md (AI instructions) Creates new skills following Anthropic's specification. When user requests a…
We split this AGENTS.md into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 46
When user requests a new workflow:
- Distinctive: line 53
Update project README.md: - Add new workflow to the list
- Distinctive: line 57
After creating any workflow, ALWAYS update the project README.md:
- Distinctive: line 59
Add new workflow to the workflow list table
- Distinctive: line 120
Before completing workflow creation:
≈ “Common” is an estimate: at least 1 in 5 of the 36,368 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AI Workflow Collection - AI Instructions
This is the master repository for curated Claude Code workflow collections.
Language Guidelines
IMPORTANT: All AGENTS.md files in this project MUST be written in English only. This ensures consistent AI behavior across different language contexts.
AGENTS.md- English only (AI instructions)README.md- English (primary)README_cn.md- Chinese translation (optional)
Project Structure
ai-workflow/
├── AGENTS.md # This file (AI context)
├── README.md # User documentation (update after creating workflows)
├── .claude/skills/ # Shared skills for this project
│ ├── workflow-creator/ # Create new workflows
│ └── skill-creator/ # Create new skills
├── workflows/
│ ├── content-creator-workflow/ # Example workflow
│ ├── marketing-pro-workflow/ # Example workflow
│ └── <new-workflow>/ # New workflows created hereAvailable Skills
workflow-creator
Trigger: "create workflow", "new workflow", "set up workflow", "build a xxx-workflow"
Creates complete workflow directories with:
- Multi-AI tool support (.claude, .codex, .cursor, .opencode, .agents, .kilocode, .roo, .goose, .gemini, .agent, .github, skills, .factory, .windsurf)
- Curated skills downloaded from GitHub
- README.md (user documentation)
- AGENTS.md (AI instructions)
skill-creator
Trigger: "create skill", "new skill", "build a skill"
Creates new skills following Anthropic's specification.
Workflow Creation Process
When user requests a new workflow:
- Create directory using
workflow-creator/scripts/create_workflow.py(default target:workflows/) - Select skills from
workflow-creator/references/skill-sources.md - Download skills using
workflow-creator/scripts/download_skill.py - Generate README.md for users
- Generate AGENTS.md for AI
- Update project README.md - Add new workflow to the list
Post-Creation Tasks
After creating any workflow, ALWAYS update the project README.md:
- Add new workflow to the workflow list table
- Update workflow count
- Add brief description
Output Standards
Workflow Directory Structure
workflows/<workflow name>/
├── README.md # User documentation
├── AGENTS.md # AI instructions (concise, <500 lines)
├── .claude/
│ ├── settings.json
│ └── skills/ # Downloaded skills (primary storage)
├── .codex/
│ └── skills -> ../.claude/skills
├── .cursor/
│ └── skills -> ../.claude/skills
├── .opencode/
│ └── skill -> ../.claude/skills
├── .agents/
│ └── skills -> ../.claude/skills
├── .kilocode/
│ └── skills -> ../.claude/skills
├── .roo/
│ └── skills -> ../.claude/skills
├── .goose/
│ └── skills -> ../.claude/skills
├── .gemini/
│ └── skills -> ../.claude/skills
├── .agent/
│ └── skills -> ../.claude/skills
├── .github/
│ └── skills -> ../.claude/skills
├── skills -> .claude/skills
├── .factory/
│ └── skills -> ../.claude/skills
└── .windsurf/
└── skills -> ../.claude/skillsREADME.md Content
- Workflow purpose and target users
- Installed skills table
- Quick start guide
- Usage examples
AGENTS.md Content
- Workflow overview
- Available skills with triggers
- Recommended workflow steps
- Skill usage guidelines
Skill Sources Priority
- Anthropic Official:
github.com/anthropics/skills - Curated Community: Listed in
skill-sources.md - Dynamic Search: GitHub search as fallback
Quality Checklist
Before completing workflow creation:
- All skills have valid SKILL.md
- Symlinks work for all AI tools
- README.md is complete
- AGENTS.md is concise and actionable
- Project README.md updated with new workflow
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 28,693 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository7 other files it gives its agents.
- add-new-skills-to-workflowSkill
- legacy-to-ai-readySkill
- remove-old-skills-from-workflowSkill
- resource-scoutSkill
- skill-creatorSkill
- skill-downloaderSkill
- workflow-creatorSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

