sciagent-skill-creatorlisted

Scaffold a new SciAgent-Skills entry. Picks pipeline/toolkit/database/guide template, creates skills/{category}/{name}/SKILL.md with valid frontmatter, appends the registry.yaml entry, runs validation. Enforces name uniqueness, kebab-case, description keyword rules, schema rules from CLAUDE.md. TRIGGER when user says (any language): "add a SciAgent skill", "add a skill for <X>", "create new skill", "create a SKILL.md for <X>", "scaffold a skill", "new skill entry", "register a skill", "신규 skill 추가", "스킬 만들어줘", "스킬 생성", "skill 만들어", or any request to add a new SKILL.md to this repo. ALWAYS invoke this skill BEFORE writing to skills/ or registry.yaml. DO NOT TRIGGER when: editing existing entry's content (just edit the file directly); migrating an existing entry (read CLAUDE.md "Migrating from Existing Entries" first); only updating registry.yaml without creating a new SKILL.md.
jaechang-hits/SciAgent-Skills · ★ 193 · Data & Documents · score 76

Install: claude install-skill jaechang-hits/SciAgent-Skills

# SciAgent Skill Creator Repo-local scaffolder for `skills/` entries. Mechanizes the boilerplate from `CLAUDE.md` Steps 1, 2, 4, 5, 6 so authoring effort stays on *content* (When to Use, Workflow, Recipes, References) and not on field plumbing. ## When to invoke this skill - User asks for a new SciAgent skill entry on a specific tool, library, database, or guide topic - User invokes `/sciagent-skill-creator` directly - The agent is about to hand-edit `registry.yaml` and create a `skills/<cat>/<name>/SKILL.md` from scratch — use this instead Do **not** invoke for: - Editing an existing entry's content (just edit the file) - Migrating an existing entry (read `CLAUDE.md` "Migrating from Existing Entries" first — the scaffolder generates a skeleton, but migration requires content judgment) - Updating `registry.yaml` only (use a normal edit) ## What you need to collect from the user Before calling the scaffold script, gather these — in conversation, not via flags hidden from the user: 1. **Topic** — concrete tool/library/concept name. Reject vague topics ("ML stuff") with a clarifying question. 2. **Sub-type** — `pipeline` | `toolkit` | `database` | `guide`. Use the decision rule from CLAUDE.md Step 1b. If unsure, ask the user. 3. **Category** — primary category directory. List the table from `CLAUDE.md` Step 2 if the user is unsure. 4. **Entry name** — kebab-case slug. Convention: `{tool-name}-{purpose}` (e.g., `pydeseq2-differential-expression`). Confirm with the user. 5.