Skill designer: generate new skills from structured idea specs
Turn a JSON skill idea into a complete skill directory — a prompt builder plus validation — that powers the repo's auto-generation pipeline for trading-analysis skills
- What
- Turn a JSON skill idea into a complete skill directory — a prompt builder plus validation — that powers the repo's auto-generation pipeline for trading-analysis skills
- Cost
- Free
- Needs
- Python 3.9+, the claude-trading-skills repo cloned locally (the prompt builder and references/ files live there), and the Claude CLI. This skill generates OTHER skills for a trading-oriented repo — it places no trades itself.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
⚠️ **Trading warning / Avertissement trading** — this skill powers an auto-generation pipeline for trading-analysis skills; trading with real money is risky, nothing in this skill places trades, and Skill Harbor provides no financial advice. Curated by Skill Harbor — @tradermonty's meta skill: it doesn't trade itself, it designs other skills. It turns a structured skill-idea specification (title, description, category like trading-analysis) into a complete Claude CLI prompt that instructs Claude to create a full skill directory — SKILL.md with YAML frontmatter, reference documents, helper scripts, and test scaffolding — following the repository conventions. The workflow: prepare the idea JSON (--idea-json) with a normalized skill name (--skill-name); run the prompt builder (python3 skills/skill-designer/scripts/build_design_prompt.py), which loads the idea, reads the three reference files (skill-structure-guide, quality-checklist with the dual-axis reviewer's 5-category 100-point rubric, skill-template), and lists existing skills to prevent duplication; pipe the resulting prompt into claude -p (--allowedTools Read,Edit,Write,Glob,Grep); then validate the output (frontmatter, directory structure, dual-axis-skill-reviewer score threshold). Honest caveats: it needs Python 3.9+, the repo's references/ files in place, and the Claude CLI; the generated skills land in the claude-trading-skills repo — expect them to be trading-oriented by design, and review generated code before running anything that touches real money. MIT licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Install
Prerequisites: Python 3.9+, the claude-trading-skills repo cloned locally (the prompt builder and references/ files live there), and the Claude CLI. This skill generates OTHER skills for a trading-oriented repo — it places no trades itself. Install "Skill designer: generate new skills from structured idea specs" for me. It gives my agent @tradermonty's skill-designer: prepare a JSON idea spec (title, description, category), run the prompt builder script (python3 skills/skill-designer/scripts/build_design_prompt.py --idea-json --skill-name --project-root), which loads the idea plus the structure guide, quality checklist and template and guards against duplicates, pipe the output into claude -p --allowedTools Read,Edit,Write,Glob,Grep, then validate the generated skill directory. MIT-licensed. Repository: https://github.com/tradermonty/claude-trading-skills/blob/main/skills/skill-designer/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "skill-designer". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. clone the repo locally so the scripts/ and references/ exist; prepare your idea JSON — review anything generated before running it against real money). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.