Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/tomzx/agents/create-cli-designnpx skills add tomzx/agents --skill create-cli-designgit clone --depth 1 https://github.com/tomzx/agentsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00072 | $0.01805 |
| Opus 5 | $0.00036 | $0.00903 |
| Sonnet 5 | $0.00014 | $0.00361 |
| Haiku 4.5 | $0.00007 | $0.00180 |
Grade A, and why
create-cli-design scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create CLI Design
Designs the command-line interface for a feature as a companion artifact to the requirements document, settling the commands, subcommands, options, arguments, exit codes, output behavior, help text, and example sessions before implementation begins.
For a CLI, the interface is the product: the user experiences the feature entirely through command syntax, help text, output, and errors.
This skill is the terminal counterpart of /create-mockups, which serves features with a graphical interface.
Without this step, CLI work starts with no shared picture of the command surface, so naming, flags, and output are improvised during implementation and reworked in review.
Prerequisites
- Apply the shared SDLC conventions in
skills/sdlc/references/shared.md. - If no argument is provided, locate the feature directory under
.sdlc/features/whose frontmatterissuefield references$ISSUE_NUMBER. .sdlc/features/N-<slug>/requirements.md(a draft is acceptable; this skill typically runs delegated from/create-requirements), or a feature brief provided in context or as a file path ($1)- The repository's existing CLI conventions, when one exists: detect the framework and match its command naming, flag style, and output conventions instead of inventing new ones
Steps
- Detect revision mode: if
review-requirements.mdexists besiderequirements.mdwithverdict: changes-requestedand its findings concerncli-design.md, read the design and the findings together, amend the design minimally to resolve each finding (preserving content the findings did not challenge), and set frontmatterstatus: in-review. Skip the fresh drafting below. - Read the requirements (or brief) and apply any style rules in
.sdlc/context/conventions.md. - Survey the existing CLI, if any: framework, command naming, flag style, output conventions, exit code usage. When the project has no CLI yet, fix the baseline conventions explicitly (e.g., GNU-style long options, kebab-case names, verb-first commands) and record them as the design principles.
- Determine whether the feature has a CLI surface: does it add or change commands, options, or command output? If not, emit
verdict: skippedand write no artifact. - Derive the command tree from the functional requirements: every user-facing FR maps to at least one command, subcommand, or option; one command per user goal; group related operations under noun subcommands.
- For each command, specify: a synopsis, a description, positional arguments (name, required, description), options (short and long form, value type, default, env-var fallback), at least one usage example, and its specific error cases.
- Define the cross-command surface: global options, environment variables, config file and precedence (flags over env over config over defaults), exit codes, the stdout/stderr split, machine-readable output (
--json) when scripts must consume the output, and TTY-aware behavior (color, progress, and tables only on a TTY, honoringNO_COLOR). - Design user error protection: an actionable error message format (
error: <message>plus a hint), confirmation prompts on destructive actions with--yes/--forceto skip, and--dry-runwhere meaningful. - Write the help text for the root command and one representative subcommand.
- Write example sessions as fenced
consoletranscripts demonstrating the intended experience: a happy path, at least one error path, and (for destructive or interactive CLIs) a confirmation exchange plus a piping example. These transcripts are the artifact's centerpiece: a reader should be able to feel what using the CLI will be like before it exists. - Feed implied changes back: options or commands that surface a missing requirement go into
requirements.mdas open questions or new FRs, and unresolved interface decisions become open questions here. - Write the output to
.sdlc/features/N-<slug>/cli-design.md, creating the directory if it does not exist, with frontmatterstatus: draft.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 99 lines · 72 tokens per session scan A 3eece73de35d
create-cli-design is a skill published in the GitHub repository tomzx/agents (5 stars, last pushed 4d ago), licensed MIT. It adds 72 tokens to every session and 1,805 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…