skill2-create
/skill2-create
Use when creating, updating, splitting, merging, or restructuring an agent skill or skill library.
- Author
- MisterBrookT ↗
- Repository
- skill2 ↗
- Licence
- MIT
- Last repository change
- 24.07.2026
Assessment and security
- Files in this package
- 7
- Scripts
- 4
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 24.07.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
- 4 scripts in this package. Review them before installing, as Claude may execute them.
Scripts: scripts/_runtime/skill2/__init__.py, scripts/_runtime/skill2/cli.py, scripts/_runtime/skill2/models.py, scripts/_runtime/skill2/scaffold.py
What it does and when to use it
Use when creating, updating, splitting, merging, or restructuring an agent skill or skill library.
Original author description
Use when creating, updating, splitting, merging, or restructuring an agent skill or skill library.
How it works
- Scope
- Authoring
- SKILL.md
- Resources
- Commands
- Library decisions
- Common mistakes
- Output
SKILL.md preview
--- name: skill2-create description: "Use when creating, updating, splitting, merging, or restructuring an agent skill or skill library." --- # Create Agent Skills Create the smallest reusable instruction set that changes agent behavior without duplicating existing ownership. ## Scope | Signal | Landing | | --- | --- | | Distinct reusable trigger and standalone workflow | Top-level `skills/<name>/SKILL.md` | | More depth under the same trigger and workflow | `references/` under the owning skill | | Repo-bound paths, commands, or conventions | Project-local skill | | One-off answer or pure mechanical rule | No skill → script / validator / docs / lint / CI | When uncertain, prefer `references/` or project-local placement over expanding the top-level namespace. ## Authoring 1. Define user trigger, desired behavior, output, and hard constraints. 2. Find existing ownership before creating a new top-level skill. 3. Choose scope and name from user intent, not implementation or one harness. 4. Write minimal `SKILL.md`; add resources only when they reduce noise or make execution deterministic. 5. Apply changes only when requested. Run repository validator when available. ## `SKILL.md` - Plain Markdown with YAML frontmatter - `name`: kebab-case; match directory name - `description`: trigger conditions only; do not summarize workflow - Body: principles, decisions, constraints, and output contract - Keep instructions terse; explain only non-obvious choices - Use relative paths; never embed secrets or accidental machine-local paths - Preserve existing user rules unless change is required ## Resources | Resource | Use | | --- | --- | | `references/` | Heavy detail loaded under the same trigger | | `scripts/` | Deterministic, repeated, or fragile execution | | `assets/` | Files copied or transformed into deliverables | Create only needed directories. Keep references one hop from `SKILL.md`. Do not nest discoverable skills. Do not create optional metadata unless the target repository or distribution format requires it. ## Commands Skill-owned scaffold: ```bash uv run --script <skill-dir>/scripts/run -- scaffold skill <name> ``` ## Library decisions - Split when users invoke workflows independently or ownership diverges - Merge when triggers overlap, shared rules dominate, and one owner can maintain both - Keep shared content with the narrowest owner serving every consumer - Avoid duplicate instructions across sibling skills - Keep top-level namespace flat and minimal ## Common mistakes | Mistake | Better choice | | --- | --- | | Workflow summary inside `description` | Trigger-only description; workflow in body | | New skill for a long table under the same trigger | `references/` under the owner | | Global skill for one repo's conventions | Project-local skill or repo instructions | | Skill for a regex-enforceable rule | Validator, lint, or CI | | Empty resource directories | Create resources only when used | | Harness-specific instructions in shared behavior | Harness-neutral rule or adapter-specific helper | ## Output 1. Scope decision: top-level / reference / project-local / no skill 2. Skill topology and ownership boundary 3. Created or updated files 4. Deterministic validation result, when available
Files in this package (7)
- SKILL.md 3 KB
- scripts/.runtime-manifest.json 1 KB
- scripts/run 1 KB
- scripts/_runtime/skill2/__init__.py 1 KB
- scripts/_runtime/skill2/cli.py 13 KB
- scripts/_runtime/skill2/models.py 7 KB
- scripts/_runtime/skill2/scaffold.py 1 KB
Versions
The database has no other skills named skill2-create.
Database snapshot: 13.11.2026 · package version 532e436d9d91 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 86skill-library-merge-dedupe
Use when merging two skill libraries for OpenCode. - 100skill-creator
Create, install, or update skills in the workspace. Use when (1) installing a skill from a URL or remote source, (2) creating a new skill from scratch, (3) updating or restructuring existing skills. Always use this skill for any skill installation or creation task. - 100skill-creator
Create, install, or update skills in the workspace. Use when (1) installing a skill from a URL or remote source, (2) creating a new skill from scratch, (3) updating or restructuring existing skills. Always use this skill for any skill installation or creation task. - 90skill-library-creator
Create a portable skill library with its own family manager, guided setup, templates and link tools. Use for "create a skill library" or "set up a skill pool"; individual skills belong to the generated manager. - 90pdf-tools
Use when creating, merging, splitting, filling, or extracting from a PDF — library-per-task, output validated. Triggers on 'merge these PDFs', 'fill this PDF form', 'split the PDF', 'create a PDF'. - 90create-skill
Create or restructure agent skills — frontmatter, description routing, progressive disclosure. Use when asked to create, write, or build a new skill, or "make a skill for X". Don't use for eval-driven trigger optimization (skill-creator) or one-off prose edits to a skill.