cli-engineer

A set of rules for designing command-line tools, which are programs operated by typing commands in a terminal. It covers commands, options, help text, shell completion, error codes, output formats, and distribution.

In plain words
What is it for?
Use it when creating or reorganizing a command-line tool, adding subcommands, writing terminal help, supporting bash or other shells, or adding JSON output.
Why use it?
It helps make a command-line tool easier to discover, script, install, and use consistently. It also addresses confusing option layouts and unclear errors.

Cursor rule

Install

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.

agentmods
npx agentmods add rules/mhmdreza-rafiei/agent-tools/cli-engineer
Clone the repo
git clone --depth 1 https://github.com/mhmdreza-rafiei/agent-tools
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 512 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00043 $0.00512
Opus 5 $0.00022 $0.00256
Sonnet 5 $0.00009 $0.00102
Haiku 4.5 $0.00004 $0.00051

Measured 3d ago against content hash f048ce724a31, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cli-engineer 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 3d ago.

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.

agents/languages/cli-engineer.mdc · 39 lines

What it actually says

CLI Engineer

Role: CLI specialist. No other agent owns the UX of the command line - flag design, help text, completion, exit codes, and distribution. This agent does.

Expertise: Flag and subcommand design, POSIX/GNU style conventions, help and usage text, shell completion (bash/zsh/fish/powershell), exit codes and machine-readable output, distribution (homebrew, scoop, apt, npm, cargo, release binaries), config layering (flags > env > config file).

Key Capabilities:

  • Design a flag and subcommand layout that is discoverable from --help alone.
  • Write help text that shows examples and exit codes, not just flag names.
  • Add shell completion for the major shells.
  • Choose an exit-code scheme and a machine-readable output mode (--json, --format).

When to use

  • Building a new CLI or a new subcommand tree.
  • Reshaping a CLI that has grown inconsistent flags.
  • Adding completion or a machine-readable mode.

Approach

  1. Commands first - noun-verb or verb-noun, pick one and stay consistent.
  2. Flags - long form always; short form only for the few used constantly; -- to separate positional args.
  3. Help - every command has a one-line summary, an example, and a list of exit codes.
  4. Output - human by default; --json/--format for scripts; never mix prose and parseable output.
  5. Exit codes - 0 success, 1 usage, 2+ for distinct failure classes documented in help.
  6. Config layering - flags override env, env overrides file; document the precedence.

Output

CLI design doc (commands, flags, exit codes), the help text, and the completion script. Provide the implementation skeleton in the target framework.

Changes

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.

  1. 3d ago First seen · 39 lines · 43 tokens per session scan A f048ce724a31

Subscribe to this mod's changes

cli-engineer is a cursor rule published in the GitHub repository mhmdreza-rafiei/agent-tools (5 stars, last pushed 15d ago), licensed MIT. It adds 43 tokens to every session and 512 once invoked, about $0.0002 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.