Command creator

Skill svyatov/agent-toolkit/plugins/command-creator/skills/command-creator

Personal agent toolkit — reusable skills and plugins for Claude Code, Cursor, Codex, and Gemini CLI

Install

npx -y skills add svyatov/agent-toolkit --skill command-creator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Guides creation of Claude Code slash commands. Use when users ask to "create a command", "make a slash command", "add a command", want to automate a repetitive workflow ("I keep doing X, can we make a command for it?"), or want to document a consistent process as a reusable slash command — project-level or global.

SKILL.md

3.4 KB, as published. Nobody here has run it

Command Creator

Create Claude Code slash commands — reusable workflows invoked with /command-name. A slash command is a markdown file in .claude/commands/ (project-level) or ~/.claude/commands/ (global) that expands into a prompt when invoked. Good candidates: repetitive workflows (code review, PR submission, CI fixing), multi-step processes that need consistency, agent delegation patterns, and project-specific automation.

Bundled Resources

Load these as needed during the workflow below:

references/patterns.md — command patterns (workflow automation, iterative fixing, agent delegation, simple execution)
references/examples.md — real command examples with full source (submit-stack, ensure-ci, create-implementation-plan)
references/best-practices.md — template structure, writing guidelines, quality checklist, common pitfalls

Workflow

Step 1: Determine Location

Check git rev-parse --is-inside-work-tree 2>/dev/null: inside a git repo → project-level .claude/commands/; otherwise → global ~/.claude/commands/. An explicit user request ("global", "user-level", "project") overrides the default. Report the chosen location before proceeding.

Step 2: Identify Command Pattern

Load references/patterns.md to understand the available patterns:

Workflow Automation — Analyze → Act → Report (e.g., submit-stack)
Iterative Fixing — Run → Parse → Fix → Repeat (e.g., ensure-ci)
Agent Delegation — Context → Delegate → Iterate (e.g., create-implementation-plan)
Simple Execution — Run command with args (e.g., codex-review)

Infer the best pattern from the user's description. Only ask if genuinely ambiguous.

Step 3: Gather Command Information

Interview the user for whatever isn't already clear from their request:

Name and purpose — kebab-case filename, matching the invocation (my-command.md → /my-command); a concise, action-oriented description for /help output.
Arguments — required or optional, and what they represent. If the command takes arguments, add argument-hint: to the frontmatter (<angle-brackets> for required, [square-brackets] for optional).
Workflow steps — the order of actions, tools/commands to use, how to handle results, success criteria, and error handling.
Tool guidance — specific agents/tools to use or avoid, files to read for context.

Step 4: Generate the Command

Load references/best-practices.md for the template structure, writing style guidelines, and quality checklist. Key principles: imperative verb-first instructions, explicit and specific actions, stated expected outcomes, concrete examples, clear error handling.

Step 5: Create the File

Ensure the target directory exists, write the command file, then confirm with the user: report the file location, summarize what the command does, and show how to invoke it (/command-name [arguments]).

Step 6: Test and Iterate (Optional)

Suggest testing with /command-name [arguments] and iterate on feedback.

Gives 0 of the 12 instructions most automation workflows skills give

Counted across 745 of the 1,008 authors here whose files we hold, read 2026-08-06

write conventional commit messagesin 36 of 745, across 35 files
delete branches after mergein 30 of 745, across 21 files
make atomic commitsin 25 of 745, across 15 files
write minimal code to pass testsin 22 of 745, across 10 files
run tests before committingin 21 of 745, across 13 files
re-snapshot after navigation or DOM changesin 21 of 745, across 13 files
use try-catch for error handlingin 20 of 745, across 6 files
write tests before implementationin 20 of 745, across 8 files
configure branch protection rulesin 19 of 745, across 5 files
explain the why in commit messagesin 19 of 745, across 9 files
refactor code while tests remain greenin 19 of 745, across 6 files
Interact with elements using refsin 19 of 745, across 11 files

Said here and by no other author read

load the referenced resource files as needed
determine command location from the git context
let explicit user location requests override the default
report the chosen file location before proceeding
identify the best command pattern from the user description
interview the user for missing command details

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.