click-llm
A tiny package that makes Click CLIs self-describing for LLMs.
click-llm adds an llm command to your CLI so models (and humans) can
discover commands, options, arguments, and command structure.
The command supports one flag:
--json: output machine-readable catalog JSON- default output: concise text catalog
Install
pip install click-llm
After install, llm is auto-enabled for Click CLIs in that Python environment.
Usage
mycli llm
mycli llm --json
Example CLI
import click
@click.group(help="Acme operations CLI.")
def acme() -> None:
pass
@acme.group(help="Deployment workflows.")
def deploy() -> None:
pass
@deploy.command(help="Roll out a service release.")
@click.argument("service")
@click.option(
"--env",
type=click.Choice(["staging", "prod"]),
default="staging",
show_default=True,
)
@click.option("--dry-run", is_flag=True, help="Plan only; do not execute.")
def release(service: str, env: str, dry_run: bool) -> None:
click.echo(f"release {service} to {env} (dry_run={dry_run})")
@acme.command(help="Show system health.")
@click.option("--json", "as_json", is_flag=True, help="Machine-readable health.")
def health(as_json: bool) -> None:
click.echo("ok")
if __name__ == "__main__":
acme()
Example output:
$ acme llm
catalog_version: 1
root_command: acme
commands: 4
### acme deploy release
summary: Roll out a service release.
usage: acme deploy release [OPTIONS] SERVICE
params:
- option `--env`, type=choice, required=False, default="staging"
- option `--dry-run`, type=boolean, required=False, default=false, is_flag=True
- argument `service`, type=text, required=True, nargs=1, default=null
$ acme llm --json
The JSON output includes the full command tree and a flattened commands list,
which is useful for prompt context, tool routing, and command planning.
Explicit integration
If you prefer explicit wiring in the host CLI:
import click
from click_llm import attach
@click.group()
def cli():
pass
attach(cli)
Notes
Auto mode uses a .pth startup hook plus a Click monkeypatch
(click.Group.get_command and click.Group.list_commands).
Metadata
Release files for click-llm 0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| click_llm-0.2.tar.gz | 9.5 kB | Without provenance | Details for click_llm-0.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| click_llm-0.2-py3-none-any.whl | Python 3 | none | any | Details for click_llm-0.2-py3-none-any.whl |
Total release size: 16.3 kB
Release files / click_llm-0.2.tar.gz
| Download URL | click_llm-0.2.tar.gz |
|---|---|
| Size | 9.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0842d9d528ef7ab30f7973b5559333317b592e3af61ead2eeea9bfc50f2aa2eb
|
|
BLAKE2b-256 checksum How to use checksums |
bc6bdff793e3b4c9bfac543fe1088877140851e8bfa46e273abbdac550c7b773
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.6.0
|
Release files / click_llm-0.2-py3-none-any.whl
| Download URL | click_llm-0.2-py3-none-any.whl |
|---|---|
| Size | 6.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
0e5f10798d4e3fa8b975b6000e01030bd9de48d8c50ec7b65a2679fc392b6013
|
|
BLAKE2b-256 checksum How to use checksums |
c8568e39512f8462db8485f5c95573371c8b836fcfa30ef8a40668ffe3e5fd0b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.6.0
|
Security for click-llm 0.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.