create-cli
Design CLI interfaces (commands, flags, I/O, errors, config) and produce implementation roadmaps with framework choice and shipping strategy.
Skill by therealtinhtute
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/therealtinhtute-create-cli
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/create-cli
What's inside
- Skill name
create-cli- Description
- Design CLI interfaces (commands, flags, I/O, errors, config) and produce implementation roadmaps with framework choice and shipping strategy. Handles greenfield and retrofit. Invoke only from a mono-mode playbook step or when the user asks for it by name.
- License
- MIT
- Compatibility
- Designed for Claude Code
- Metadata
- version: 1.0.0
- Instructions
- 130 lines in SKILL.md · view on GitHub
Files (6)
- SKILL.md
- cli-guidelines.md · references/
- examples.md · references/
- framework-matrix.md · references/
- shipping-checklist.md · references/
- spec-template.md · references/
SKILL.md (first lines)
Detect the mode first, then interview.
<role>
Act as a CLI architect. Design command-line interfaces that are human-first, script-friendly,
and shippable. Produce specs and implementation roadmaps — not code. Pick the right framework
for the constraints, plan distribution, and hand off to implementation skills.
</role>
<context>
## When to Use
- New CLI (greenfield), or an existing script formalized into a CLI (retrofit)
- CLI framework choice (Go/Rust/Node/Bash), distribution, and packaging
## Defer To Instead
- `think` — general architecture decisions not specific to CLIs
- `work` — actual implementation after spec is approved
- `check` — auditing CLI code quality and security
</context>
<instructions>
## Mode Detection
Determine mode from user input:
**Greenfield** — user describes what the CLI should do, no existing code referenced.
**Retrofit** — user points to an existing script/binary, wants to formalize the interface.
If ambiguous, ask.
---
## Greenfield Mode
### Phase 1: Fast Clarify
Ask these via the agent's question tool (`AskUserQuestion` on Claude Code) (batch max 4, recommended option first):
1. **Command name** — what users type. Short, memorable, no hyphens if possible.
2. **One-liner** — what it does in ≤10 words.
3. **User type** — developer, ops, end-user, or CI/automation.
4. **Language/framework** — recommend from `references/framework-matrix.md` (distribution, team expertise, performance, existing deps).
Then ask:
5. **Input sources** — stdin, files, args, env, API?
6. **Output contract** — human text, JSON, both (detect TTY)?
7. **Interactivity** — fully interactive, `--no-input` mode, or non-interactive only?
8. **Config model** — flags only, env vars, config file, or layered?
### Phase 2: Design Spec
Fill `references/spec-template.md` and enforce every convention in `references/cli-guidelines.md` (standard flags, exit codes, stdout/stderr, `NO_COLOR`, config precedence, XDG). Also decide:
- Command tree at most 2 levels deep unless justified; one subcommand order (noun-verb or verb-noun), used consistently
- Every flag: long form, short form if warranted, type, default, description
- `-f`/`--force` skips confirmations on dangerous operations only
- Per command: human vs JSON output; error pattern, codes, recovery hints
- Config file format and location, shell completions, SIGINT/SIGTERM handling, target platforms
### Phase 3: Implementation Roadmap
After spec approval, produce:
1. **Framework choice** with rationale (`framework-matrix.md`)
2. **Project structure** — directories and key files
3. **Ordered task list**, one PR-sized unit each: scaffold + arg parsing + `--help`/`--version`; one task per subcommand; config loading; human + JSON output; errors and exit codes; completions; unit + integration tests; distribution
4. **Shipping plan** — how it reaches users (`shipping-checklist.md`)
---
## Retrofit Mode
### Phase 1: Extract Current Interface
Read the existing script/code and document its **as-is interface** in spec format: arguments, env vars read, output format and destination, exit codes, config files.
### Phase 2: Gap Analysis
Compare as-is against `cli-guidelines.md`:
| Convention | Current | Target | Breaking? |
|---|---|---|---|
| `--help` | missing | add | no |
| exit codes | always 0 or 1 | standard set | yes |
| ... | ... | ... | ... |
Flag breaking changes explicitly. Ask user which breaks are acceptable.
### Phase 3: Redesign Spec
Produce the target spec (greenfield Phase 2 format), marking what stays (backwards-compatible), what changes (with migration notes), and what is new.
### Phase 4: Migration Roadmap
Like greenfield Phase 3, ordered to minimize breakage: non-breaking additions (new flags, help text) → deprecation warnings → breaking changes with a version bump.
---
## Output Format
Save to: `.harness/plans/active/cli-{name}.md`, one plan with a `## Spec` and a `## Roadmap` section. IF another plan is already active under `.harness/plans/active/` → append `## CLI Spec` and `## CLI Roadmap` sections to it instead of creating a new file.
These integrate with `/brainstorm → /to-plan → /work` workflow.
Frontmatter:
```yaml
---
title: CLI Spec — {name}
description: {one-liner}
status: draft
created: {date}
tags: [cli, {language}]
---
```
See `references/examples.md` for sample spec and roadmap outputs.
</instructions>
<references>From therealtinhtute/mono-harness, MIT.
More from therealtinhtute/mono-harness
- 0.624score
turbo-mono-platform
therealtinhtute
Full-stack TypeScript monorepo specialist: Turborepo, Next.js, Hono, tRPC, Drizzle, Postgres.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
librarian
therealtinhtute
GitHub code research via gh CLI, no cloning.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
read
therealtinhtute
Search the web and fetch URLs or PDFs as clean Markdown via a privacy-first cascade (local, gh, TinyFish/Firecrawl/Exa, opt-in proxy).
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
git
therealtinhtute
Git operations with conventional commits.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
hunt
therealtinhtute
Find the root cause before any fix: build a red-capable repro loop, rank falsifiable hypotheses, prove, fix, sweep siblings.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
think
therealtinhtute
Reason hard before building: attack premises, run named thinking lenses, compare real alternatives, give one verdict.
SkillAFreshness: A★ 5Claude Code +9 - 0.619score
retro
therealtinhtute
Session retrospective: evidence-cited, severity-ranked findings about the agent's environment, then one user-approved fix — a guard or a rerun-gated experiment.
SkillAFreshness: A★ 5Claude Code +9 - 0.619score
arena
therealtinhtute
Design bakeoff: sub-agents build one artifact several ways, a rubric picks the base, the best parts of the rest are grafted on.
SkillAFreshness: A★ 5Claude Code +9 - 0.619score
mono-mode
therealtinhtute
Entry router: match a task to one of twelve playbooks (feature, bug-fix, refactoring, perf, research, ship, ...) and run its skill steps as a todo list.
SkillAFreshness: A★ 5Claude Code +9 - 0.618score
prompt-leverage
therealtinhtute
Strengthen a raw prompt into an execution-ready instruction set for Claude Code or another AI agent.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
work
therealtinhtute
Execution orchestrator after `brainstorm` (and `to-plan` for high-risk).
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
check
therealtinhtute
Pre-commit and pre-merge gate.
SkillAFreshness: A★ 5Claude Code +9
Signals
- Stars
- 5
- Forks
- 1
- Open issues
- 1
- Last commit (this folder)
- Oct 23, 2026 (5 days ago)
- Last release
- Oct 23, 2026 (5 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 1.0.0
- License
- MIT
- Repository
- github.com/therealtinhtute/mono-harness
- Website
- harness.tinhtute.me
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from therealtinhtute
- 0.624score
turbo-mono-platform
therealtinhtute
Full-stack TypeScript monorepo specialist: Turborepo, Next.js, Hono, tRPC, Drizzle, Postgres.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
git
therealtinhtute
Git operations with conventional commits.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
librarian
therealtinhtute
GitHub code research via gh CLI, no cloning.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
think
therealtinhtute
Reason hard before building: attack premises, run named thinking lenses, compare real alternatives, give one verdict.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
hunt
therealtinhtute
Find the root cause before any fix: build a red-capable repro loop, rank falsifiable hypotheses, prove, fix, sweep siblings.
SkillAFreshness: A★ 5Claude Code +9 - 0.621score
read
therealtinhtute
Search the web and fetch URLs or PDFs as clean Markdown via a privacy-first cascade (local, gh, TinyFish/Firecrawl/Exa, opt-in proxy).
SkillAFreshness: A★ 5Claude Code +9
Cite this page
create-cli skill. Agent Mart. https://agentmart.dev/skills/therealtinhtute-create-cli (last updated 2026-10-24)