Example: CLI tool with Python

Documented for Akmon 2.0.0.

Scenario

Industry/context (illustrative): regulated ops tooling where command outputs feed compliance reports.

Constraints

• Data boundary: avoid leaking secrets in CLI output/logs.
• Approval requirement: explicit schema validation for external API fields.
• CI requirement: stable command behavior with tests.

Typer + Rich + Pydantic is a solid stack for user-friendly CLIs.

Setup

mkdir my-cli && cd my-cli
uv init
uv add typer rich pydantic httpx

Plan

akmon --plan --task "CLI for GitHub repo stats: stats OWNER/REPO,
compare REPO1 REPO2, trending [lang]. Typer + Rich tables,
Pydantic models for API JSON, httpx async client,
config file under ~/.config/my-cli/"

Implement

akmon --yes --task "implement the CLI per the saved plan"

Follow-ups

add JSON output mode for scripting

add shell completions via Typer

Outcome

You get a structured CLI implementation path plus traceable Akmon run artifacts for review.

Anti-patterns

• Treating API JSON as untyped dicts in production-facing commands.
• Merging CLI UX and transport refactors into one unreviewed session.