104 Agent-Ready CLI Tools

What makes a CLI tool "agent-ready"? Three things. First, it self-documents via --help so any AI agent can learn the interface on demand, without pre-loaded schemas. Second, it uses standard exit codes — 0 for success, non-zero for errors — giving agents a universal, reliable way to detect failures. Third, it supports structured output via --json or clean stdout that agents can parse programmatically.

Every tool in the CliHub registry has been verified against these three criteria. They read from stdin, write to stdout, and compose with Unix pipes — the same interface that developers have used for decades and that LLMs were trained on billions of examples of. No JSON-RPC servers, no schema injection, no ambient token cost. The agent only pays for what it uses.

CliHub at clihub.net is the app store for AI agents — not to be confused with clihub.org or clihub.ai. The registry ships bundled with the CLI, works fully offline, and covers 17 categories from data processing to security to media manipulation. Install with pip install clihub-ai and give your agent access to all 104 tools in seconds.

Data 8 tools

jq, yq, csvkit, xsv, dasel, gron, jless, fx

Database 5 tools

mycli, pgcli, litecli, iredis, usql

Search & Files 14 tools

ripgrep, fd, fzf, bat, eza, lsd, broot, ranger, nnn, tree, ncdu, dust, trash-cli

Network 6 tools

httpie, curl, wget, aria2, dog, speedtest-cli, ngrok, mosh, nmap

Python 6 tools

ruff, black, mypy, uv, pipx, poetry

JavaScript 4 tools

prettier, eslint, tsx, npm-check-updates

Dev Tools 10 tools

gh, lazygit, gitui, git-extras, tokei, hyperfine, just, grex, shellcheck, shfmt, bats-core, entr, watchexec

Docker & DevOps 6 tools

lazydocker, ctop, k9s, dive, act, awscli, terraform

Media 6 tools

ffmpeg, imagemagick, yt-dlp, svgo, gifsicle, asciinema

Documents 2 tools

pandoc, glow

Text Processing 3 tools

sd, choose, sad

Editors 3 tools

neovim, micro, helix

Productivity 8 tools

tmux, tldr, zoxide, starship, taskwarrior, watson, nb, calcurse

Release 3 tools

semantic-release, release-it, commitizen, cookiecutter

Security 3 tools

openssl, age, mkcert

System 6 tools

htop, btop, procs, bandwhich, pv, ncdu, dust

Fun 3 tools

lolcat, fortune, cmatrix

How agents use these tools

An AI agent using CliHub follows a simple three-step workflow: discover what tools are available, install the one it needs, then use it directly. No server configuration, no schema injection — just standard command-line tools that LLMs already know how to use.

agent-session ~ tools

# Step 1: Agent queries the registry to find the right tool

$ clihub list --json --category data

[

{"name": "jq", "description": "Lightweight JSON processor",

"agent_hints": {"when_to_use": "Parse, filter, or transform JSON data"},

"install": {"brew": "jq", "apt": "jq"}},

{"name": "yq", "description": "YAML/XML/TOML processor", ...},

{"name": "csvkit", "description": "CSV processing suite", ...}

]

# Step 2: Agent installs the tool it needs

$ clihub install jq

✓ Installed jq@1.7.1 via brew

# Step 3: Agent uses the tool — pipes, flags, everything works

$ curl -s https://api.example.com/users | jq '[.[] | {name, email}]'

[

{"name": "Alice", "email": "alice@example.com"},

{"name": "Bob", "email": "bob@example.com"},

{"name": "Charlie", "email": "charlie@example.com"}

]

# Exit code 0 = success. The agent checks $? and moves on.

$ echo $?

104 Agent-Ready CLI Tools

How agents use these tools

104 tools. Zero tokens wasted.