Creator
Skill4 starsChanged 7 months agoMITView on GitHub
Multi-branch creation skill for scaffolding new skills, rules, and workflows/procedures. Ensures all created artifacts comply with project standards from day one.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- arctan303/creator.skilltheir own version, first seen Nov 12, 2026
- CraftOS-dev/Agent-Apptheir own version, first seen Nov 8, 2026
- Victorcorcos/skillstheir own version, first seen Jun 12, 2026
- thematteroftime/agentic-md-for-dummiestheir own version, first seen May 10, 2026
- marswaveai/skillstheir own version, first seen Mar 25, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- marswaveai/skills★ 82same name
- arctan303/creator.skill★ 19same name
- CraftOS-dev/Agent-App★ 10same name
- Victorcorcos/skills★ 7same name
- thematteroftime/agentic-md-for-dummies★ 6same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- .agents/skills/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- skill-ai-dspy-validatorSkill
- Bootstrap AgentSkill
- Deep ResearchSkill
- skill-devops-github-actionsSkill
- skill-guard-governanceSkill
- skill-impact-predictorSkill
- skill-review-codeSkill
- Self ReflectionSkill
- skill-surgical-refactorerSkill
- skill-test-architectSkill
- skill-threat-modeling-expertSkill
- [Skill Name]Skill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 9 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 94
Create a skill without a flowchart
- Distinctive: line 96
Leave placeholder content ("TODO: fill in steps")
- Distinctive: line 97
Create a skill that duplicates an existing one
- Distinctive: line 137
Every rule MUST map to a cognitive branch
- Very specific: line 138
Update the Rules Registry table in COGNITIVE_TREE.md
- Distinctive: line 149
Create a rule without mapping it to a cognitive branch
- Very specific: line 150
Write advisory-only rules for critical constraints (use enforcement: deterministic)
- Distinctive: line 151
Create rules that contradict existing rules without deprecating the old one
- Distinctive: line 152
Write vague rules ("be careful with X") — always include concrete examples
- Distinctive: line 196
Write steps that say "do the right thing" without specifying what that is
- Distinctive: line 197
Create procedures longer than 200 lines (split into sub-procedures if needed)
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
🏗️ Creator — Scaffold Skills, Rules & Workflows
"Every governance artifact starts from a template, but becomes unique through substance."
This skill has 3 branches. Each branch produces a different type of governance artifact, enforcing project standards at creation time so violations are impossible.
Branch Router
flowchart TD
START["Need to create\na governance artifact"] --> Q1{"What type?"}
Q1 -->|"Reusable agent\ncapability"| SKILL["Branch A\nSkill Creator"]
Q1 -->|"Immutable\nproject standard"| RULE["Branch B\nRule Creator"]
Q1 -->|"Step-by-step\noperational guide"| WORKFLOW["Branch C\nWorkflow Creator"]
SKILL --> VALIDATE["Validate against\nrule-consistency"]
RULE --> VALIDATE
WORKFLOW --> VALIDATE
VALIDATE --> DONE["✅ Ready for review"]Branch A: Skill Creator
When to Use
- Agent discovers a repeatable capability that should be formalized
- User requests a new agent skill
- Importing a skill from an external source (e.g., skills.sh, GitHub)
Creation Protocol
- Name the skill —
skill-{name}/SKILL.md.agents/skills/skill-{name}/SKILL.md - Generate frontmatter (mandatory fields)
--- name: {Human-readable name} description: > {What this skill does, when to use it} version: 1.0.0 triggers: - {when should this skill activate} --- Write the body (mandatory sections)
Section Required Purpose Overview flowchart ✅ Visual decision path (mermaid) When to Use ✅ Trigger conditions Steps ✅ Concrete execution steps Do Not Do ✅ Explicit anti-patterns Quick Reference ⭐ Optional cheat sheet - Validate
- Has mermaid flowchart (rule-consistency §5)
- No TODO/TBD/PLACEHOLDER (rule-zero-theater)
- All content in English (rule-adaptive-language)
- Description is specific, not generic (rule-lesson-quality spirit)
Importing External Skills
flowchart TD
SOURCE["External skill\n(URL, repo, skills.sh)"] --> FETCH["Fetch/download\nskill content"]
FETCH --> ADAPT["Adapt to project\nfrontmatter format"]
ADAPT --> VALIDATE["Validate against\nproject rules"]
VALIDATE -->|"Pass"| INSTALL["Install to\n.agents/skills/"]
VALIDATE -->|"Fail"| FIX["Fix violations\nbefore installing"]Do Not Do
- Create a skill without a flowchart
- Copy a skill verbatim from another project without adapting frontmatter
- Leave placeholder content ("TODO: fill in steps")
- Create a skill that duplicates an existing one
Branch B: Rule Creator
When to Use
- A recurring pattern violation needs codification
- A lesson keeps appearing (> 3 occurrences)
- A new governance constraint is needed
Creation Protocol
- Name the rule —
rule-{name}.md.agents/rules/rule-{name}.md - Generate frontmatter (mandatory fields)
--- id: RULE-{UPPER-CASE-NAME} status: active version: 1.0.0 enforcement: deterministic | advisory cognitive_branch: evidence | mechanism | growth | hitl_trunk --- Write the body (mandatory sections)
Section Required Purpose Tagline quote ✅ One-line philosophical summary Decision flowchart ✅ Visual decision path (mermaid) Do / Do Not Do tables ✅ Explicit behaviors Anti-Patterns table ✅ ❌/✅ comparison Executable Logic ✅ Regex for automated detection - Map to COGNITIVE_TREE
- Every rule MUST map to a cognitive branch
- Update the Rules Registry table in COGNITIVE_TREE.md
- Validate
- Has YAML frontmatter with all required fields
- Has mermaid flowchart
- Has Do-Not-Do section
- Has executable logic regex
- Mapped to cognitive branch
- Listed in COGNITIVE_TREE Rules Registry
Do Not Do
- Create a rule without mapping it to a cognitive branch
- Write advisory-only rules for critical constraints (use
enforcement: deterministic) - Create rules that contradict existing rules without deprecating the old one
- Write vague rules ("be careful with X") — always include concrete examples
Branch C: Workflow/Procedure Creator
When to Use
- A multi-step process needs to be documented
- A repeatable operational task needs standardization
- An agent keeps asking "how do I do X?"
Creation Protocol
- Name the procedure —
procedure-{name}.md.agents/workflows/procedure-{name}.md - Generate frontmatter
--- description: {Short title of what this procedure does} --- Write the body (mandatory sections)
Section Required Purpose Overview flowchart ✅ Visual process path (mermaid) Prerequisites ✅ What must be true before starting Steps (numbered) ✅ Concrete, actionable steps Verification ✅ How to confirm success Do Not Do ✅ Common mistakes Troubleshooting ⭐ Optional error recovery - Validate
- Has mermaid flowchart
- Steps are numbered and actionable
- Each step has a verification check
- No TODO/TBD placeholders
Do Not Do
- Mix procedure steps with rule definitions
- Write steps that say "do the right thing" without specifying what that is
- Create procedures longer than 200 lines (split into sub-procedures if needed)
- Embed rule content that belongs in
.agents/rules/
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.

