This skill should be used when the user asks to "create a workflow", "make a slash command", "add a command", "automate a task with a command", "workflow vs plugin", "should this be a workflow or plugin", or needs guidance on Claude Code command structure, workflow design patterns, or determining appropriate scope for automation requests.
This skill helps you create effective slash commands for focused, one-shot tasks. Use it when users request workflow creation or need guidance on whether a task should be a simple workflow versus a full plugin.
/plugin marketplace add FrancisVarga/coconut-claude-code-plugins/plugin install coconut-workflow@coconut-claude-code-pluginsThis skill inherits all available tools. When active, it can use any tool Claude has access to.
This skill provides guidance for creating effective Claude Code workflows (slash commands) and determining when a request needs a simple workflow versus a full plugin.
Key concepts:
| Criterion | Workflow Appropriate |
|---|---|
| Purpose | Single, focused task |
| State | No persistent state needed |
| Execution | One-shot, runs and completes |
| Scope | Project-specific or personal |
| Complexity | Simple logic, no validation hooks |
| Integration | Uses built-in tools only |
| Reusability | Used in one project mainly |
Examples of workflows:
review-pr - Review current PR changesrun-tests - Execute test suite with coveragegenerate-docs - Create documentation for filecommit-changes - Stage and commit with messagefix-lint - Auto-fix linting issues| Criterion | Plugin Required |
|---|---|
| Commands | Multiple related commands needed |
| Automation | Needs hooks for automatic triggering |
| Autonomy | Needs agents for complex autonomous tasks |
| Knowledge | Needs skills for specialized guidance |
| Integration | External service via MCP servers |
| Scope | Reusable across multiple projects |
| Validation | Complex validation or enforcement logic |
Examples requiring plugins:
Ask these to determine scope:
If all answers are "No" → Workflow is appropriate.
.claude/commands/
├── workflow-name.md # /workflow-name command
└── category/
└── specific.md # /specific (project:category)
---
description: Brief description under 60 chars
argument-hint: [required-arg] [optional-arg]
allowed-tools: Read, Write, Grep
---
[Instructions for Claude - imperative form]
| Field | Required | Purpose |
|---|---|---|
description | Recommended | Shows in /help, under 60 chars |
argument-hint | If args used | Documents expected arguments |
allowed-tools | Recommended | Restrict tools (security) |
model | Optional | Force specific model |
Critical Rule: Write instructions FOR Claude to execute, not messages TO the user.
Correct (imperative form):
Review the code changes in the current branch.
Check for:
1. Security vulnerabilities
2. Performance issues
3. Code style violations
Report findings with file paths and line numbers.
Incorrect (user-facing):
This command will review your code.
You'll receive a report with issues found.
---
argument-hint: [file-path]
---
Review @$1 for code quality issues.
Usage: /review src/api/users.ts
---
argument-hint: [source] [target]
---
Compare @$1 with @$2 and list differences.
Usage: /compare old.js new.js
---
argument-hint: [search-terms...]
---
Search codebase for: $ARGUMENTS
Usage: /search authentication user login
Use @ to include file contents:
Review @$1 for issues. # File from argument
Review @package.json settings. # Static file reference
Only grant tools the workflow actually needs:
| Task Type | Recommended Tools |
|---|---|
| Read-only analysis | Read, Grep, Glob |
| Code modification | Read, Write, Edit |
| Git operations | Bash(git:*) |
| Testing | Bash(npm:test), Read |
| Full automation | Omit field (all tools) |
# Specific tools
allowed-tools: Read, Write, Grep
# Bash with restrictions
allowed-tools: Bash(git:*), Bash(npm:*)
# All tools (rarely needed)
allowed-tools: "*"
Before finalizing a workflow, verify:
.claude/commands/---
description: Review code changes for quality issues
allowed-tools: Read, Grep, Glob, Bash(git:*)
---
Get changed files: !`git diff --name-only HEAD~1`
For each changed file:
1. Read the file content
2. Check for security issues
3. Check for performance problems
4. Verify code style
Report issues with:
- File path and line number
- Issue description
- Suggested fix
---
description: Run tests for specific module
argument-hint: [module-path]
allowed-tools: Bash(npm:*), Read
---
Run tests: !`npm test -- $1`
Analyze results:
- List failing tests
- Identify root causes
- Suggest fixes
If all pass, confirm success.
---
description: Generate documentation for file
argument-hint: [source-file]
---
Read and analyze @$1
Generate documentation including:
- Purpose and overview
- Function/class descriptions
- Parameters and return values
- Usage examples
Write to appropriate location based on project structure.
---
description: Commit changes with conventional message
argument-hint: [type] [description]
allowed-tools: Bash(git:*)
---
Stage changes: !`git add -A`
Show staged: !`git diff --cached --stat`
Create commit with type "$1" and message "$2".
Follow conventional commits format:
- feat: new feature
- fix: bug fix
- docs: documentation
- refactor: code refactoring
Confirm commit created successfully.
Workflow not appearing in /help:
.claude/commands/.md extensionArguments not working:
$1, $2 for positional$ARGUMENTS for allTools not available:
Bash(pattern:*) for bash restrictionsFile references not loading:
@ prefix for file pathsWhen a workflow needs to leverage plugin components:
Invoke Agent:
Use the code-reviewer agent for detailed analysis.
Reference Skill:
Apply coding-standards skill guidelines.
Trigger Another Command:
After completion, run /validate-changes.
This skill provides the foundation for creating effective, well-structured workflows that maintain team consistency and follow Claude Code best practices.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.