skill-template · diff
git:20260831.099306a to git:20261021.8facda8
8 added, 4 removed. Audit A to A.
---
name: skill-template
description: Generate custom skill/command from template. Use when creating a new skill or slash command from scratch, or scaffolding a skill file with correct frontmatter.
user-invocable: true
---
# Skill Template Generator
Generate a custom skill (slash command) based on requirements.
**Usage**: `/skill-template [skill-name] [purpose]`
---
## Templates
| Purpose | Template | Features |
|---------|----------|----------|
| Workflow | Multi-step | Sequential steps |
| Generator | Creator | File creation |
| Checker | Validator | Validation rules |
| Automation | Runner | Command execution |
| Agent-backed | Delegator | Runs in a forked subagent |
---
## Process
1. **Gather Requirements**
- Skill name (lowercase, hyphenated)
- Purpose
- Arguments (if any — space-separated names)
- Workflow steps
- Context isolation (`context: fork`)?
- Agent backing (`agent:`, only with `context: fork`)?
- Tool restrictions?
2. **Select Template** based on purpose
3. **Generate File** at `.claude/skills/[name]/SKILL.md`
4. **Validate** with `/skill-check`
---
## Frontmatter Reference (official fields)
Only `description` is meaningfully required (`name` defaults to the directory
name). All other fields are optional.
```yaml
---
name: skill-name # Optional: defaults to directory name; lowercase, hyphenated
description: What it does # Recommended: shown in / menu and used for triggering
when_to_use: When to reach for it # Optional: appended to description for triggering
# (description + when_to_use truncated at 1,536 chars in listings)
user-invocable: true # Optional: default true; false hides from / menu (Skill tool still works)
- model: sonnet # Optional: fable, opus, sonnet, haiku, inherit, default, best,
- # sonnet[1m], opus[1m] (or a full model ID). NOT opusplan (session-only)
+ model: sonnet # Optional: any /model value (sonnet, opus, haiku, fable, best,
+ # default, opusplan, sonnet[1m], opus[1m], a full model ID) or inherit
effort: medium # Optional: low, medium, high, xhigh, max
allowed-tools: # Optional: pre-approved tools. Accepts a comma-separated string
- Read # OR a YAML list — BOTH are official (list shown here as house style)
- Grep
- Glob
disallowed-tools: # Optional: explicitly blocked tools (string or YAML list)
- WebFetch
context: fork # Optional: run the skill in a subagent
agent: agent-name # Optional: subagent type when context: fork (default: general-purpose)
+ background: false # Optional (2.1.218+): with context: fork, false waits for the result;
+ # forked skills run in the background by default
argument-hint: "[issue-number]" # Optional: autocomplete hint shown after the skill name
arguments: target flags # Optional: space-separated argument NAMES for $name substitution
# (arguments "target flags" -> use $target and $flags in the body).
# NOT a structured array of name/description/required objects.
disable-model-invocation: false # Optional: bool; true also blocks scheduled-task invocation
paths: # Optional: glob patterns that limit where the skill activates
- "src/**/*.ts"
shell: bash # Optional: bash or powershell for inline !`command` blocks
hooks: # Optional: skill-scoped lifecycle hooks (same schema as settings.json)
PreToolUse:
- matcher: Bash
hooks:
- type: command
command: echo "Bash used"
Stop:
- hooks:
- type: command
command: ./scripts/verify.sh
once: true
---
```
- Non-official fields to avoid: `metadata`, `license`, `version` (not part of the
- spec). `once` is valid only inside a `hooks[]` entry, never at the top level.
+ `metadata` (a free-form map), `license`, and `compatibility` (up to 500
+ characters) are accepted but not acted on. `version` is not a skill field, and
+ Claude Code silently ignores unknown fields. `once` is valid only inside a
+ `hooks[]` entry, never at the top level.
Do not emit bare `Bash`, `Edit`, `Write`, `Agent`, `Task`, or `Skill` entries in
`allowed-tools`; those rules preapprove broad mutation or delegation. Omit the
field when the workflow should use the runtime's normal permission flow.
---
## Workflow Template Structure
```markdown
---
name: [name]
description: [What it does. Include when to use it.]
user-invocable: true
context: fork
argument-hint: "<task-description>"
# when_to_use: Use when the user asks about X or mentions Y. # Optional trigger hint
# arguments: target mode # Optional: names for $target, $mode
---
# [Skill Name]
## Workflow
### Step 1: [Name]
### Step 2: [Name]
### Step 3: [Name]
## Arguments
Uses `$ARGUMENTS` for raw input (or `$name` per the `arguments` field)
## Output
Summary when complete
```
---
## Example
```
/skill-template deploy-staging "deploy to staging"
Output: Created .claude/skills/deploy-staging/SKILL.md
Usage: /deploy-staging
```