From nodeshub-seo-skills
Scaffolds new Agent Skills-compliant skills with full directory structure and frontmatter, then auto-syncs the skills registry.
How this skill is triggered — by the user, by Claude, or both
Slash command
/nodeshub-seo-skills:skill-creatorThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Scaffold new Agent Skills-compliant skills and keep the skills registry in sync.
Scaffold new Agent Skills-compliant skills and keep the skills registry in sync.
Ask the user for:
nod-my-skill for NodesHub API skills, or my-utility for generic skills)Validate the name against Agent Skills spec:
Run the scaffolding script:
# For nod- skills (NodesHub API-based)
python3 .claude/skills/nod-nodeshub-api/scripts/create_skill.py \
--name "nod-new-skill" \
--description "Short description of what it does" \
--cost "1 token"
# For generic skills (no NodesHub dependency)
python3 .claude/skills/nod-nodeshub-api/scripts/create_skill.py \
--name "my-utility" \
--description "What this utility does" \
--type generic
Registry sync happens automatically:
sync_skills.py updates SKILL.md table and CLAUDE.md listTell the user what was created and what to do next.
.claude/skills/{name}/
├── SKILL.md # Full Agent Skills frontmatter + template
├── scripts/ # Empty — user adds implementation
└── evals/
└── evals.json # Starter eval prompts and assertions
---
name: {name}
description: |
{description with trigger phrases}
compatibility: "{environment requirements}"
metadata:
author: nodeshub
version: "0.1.0"
allowed-tools: Bash Read Write
---
All five Agent Skills spec fields — name, description, compatibility, metadata, allowed-tools.
For every new skill (with or without nod- prefix), when the skill is invoked Claude must run the banner as the first action:
python3 -c "import sys; sys.path.insert(0,'.claude/skills/nod-nodeshub-api/scripts'); from banner import print_banner; print_banner('Human-Readable Skill Name')"
Replace Human-Readable Skill Name with the skill title (e.g. "SERP Analysis", "Connect GSC").
Located in .claude/skills/nod-nodeshub-api/scripts/:
| Script | What it does |
|---|---|
create_skill.py | Scaffolds a new skill (nod- or generic) with full spec compliance, auto-runs sync |
sync_skills.py | Regenerates skills tables from nod-* directories — reads descriptions and cost from frontmatter |
sync_skills.py reads all metadata directly from each SKILL.md frontmatter. No hardcoded maps — new nod- skills are picked up automatically.
| Requirement | How this skill handles it |
|---|---|
name (lowercase, hyphens, max 64) | Validated before creation |
description (1-1024 chars, trigger phrases) | Template includes trigger phrase pattern |
compatibility | Auto-set based on skill type |
metadata (author, version) | Always included (nodeshub, 0.1.0) |
allowed-tools | Default: Bash Read Write |
evals/evals.json | Auto-generated with starter prompts |
| SKILL.md < 500 lines | Template is ~40 lines, user extends |
npx claudepluginhub senuto/nodeshub-seo-skills --plugin nodeshub-seo-skillsCreates, updates, or validates SKILL.md agent skills including frontmatter authoring, bundled resource planning, and three-phase discoverability validation.
Creates new agent skills with proper structure and progressive disclosure. Use when user wants to create, write, or build a new skill.
Creates new Claude Code agent skills with proper folder structure, SKILL.md format, and optional examples/templates. Use when adding model-invoked skills for autonomous activation.