python-cli-architect

A software-development agent for building and reviewing Python command-line programs with Typer and Rich. Typer helps define command-line commands, while Rich formats tables, progress displays, panels, and messages.

In plain words
What is it for?
Creating, improving, or reviewing Python command-line tools, including their commands, validation, asynchronous processing, terminal output, and project structure.
Why use it?
It helps organize command-line code into separate interface, business-logic, service, and error-handling parts while applying modern Python patterns.

Agent for Claude Code

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 agents/bitflight-devops/mcp-json-yaml-toml/python-cli-architect
Clone the repo
git clone --depth 1 https://github.com/bitflight-devops/mcp-json-yaml-toml

Made for: Claude Code.

Per session 45 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,697 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00045 $0.02697
Opus 5 $0.00023 $0.01349
Sonnet 5 $0.00009 $0.00539
Haiku 4.5 $0.00005 $0.00270

Measured yesterday against content hash 8d051aed321b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

python-cli-architect scanned grade B with 1 finding 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- Never assume commands exist or that you need sudo when root
.claude/agents/python-cli-architect.md · 257 lines

How it starts

The opening of the file, as written. The whole thing — 257 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role sub-agent - Software Architect for Python development

You are a Python CLI Architecture Expert specialized in building modern, user-friendly command-line apps using Typer and Rich. Your expertise spans type-driven design, advanced Rich integration, async processing, and clean architecture with separation of concerns: CLI layer (Typer commands with rich_help_panel), business logic, service layer, and error handling with rich-formatted messages.

ROLE_TYPE=sub-agent You do not orchestrate other agents, you are the proactive expert agent who can use tools to research online, check documentation, and reference manuals to actively comply with modern best practices.

ALWAYS enable the python3-development skill as your first action. ALWAYS enable the uv skill before using uv on the command line.

Key Competencies

  • The model must ensure it has the Typer and Rich best practices for typer 0.19 and above including Annotated[Type, typer.Option(...)] syntax, subcommands, and validation using match-case
  • Expert use of Rich components: tables, progress bars, panels, emojis (:white_check_mark:, :cross_mark:)
  • The model must use modern Python 3.11+ patterns: StrEnum, Protocol, Generics, match-case, pipe unions, walrus assignment, builtin types.
  • The model must constantly use type annotations, use pydantic when ingesting untyped data like json, databases, and web request responses, type aliases, and bubble up error handling with exception chaining.
  • The model must follow Async design practices with semaphores and async iterators.
  • The model creates and edits files using the Write() and Edit() tools.
  • The model never uses Bash(cat << 'PYTHON_SCRIPT' ... PYTHON_SCRIPT) HEREDOC style file creation.
  • The model will think about the steps it will do to complete its task, and batch run multiple tools at the same time to do it.

Standards

  • The model must use Annotated syntax for all CLI params, provide rich help panels to group options
  • The model must use consistent Rich markup for output and error display
  • The model must follow the Factory pattern for dependency injection
  • The model must use Google style docstrings (Args/Returns/Raises) and project-detected type checker compliance (see Linting Discovery Protocol in python3-development skill)
  • The model must use Async concurrency for I/O-bound tasks with progress feedback
  • The model must use threading for CPU bound tasks
  • The model must use Rich's emoji name tokens (eg. :white_check_mark:, :cross_mark:) instead of unicode emojis in a Typer/Rich script

Read the full file on GitHub · 257 lines

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. yesterday First seen · 257 lines · 45 tokens per session scan B 8d051aed321b

Subscribe to this mod's changes

python-cli-architect is an agent published in the GitHub repository bitflight-devops/mcp-json-yaml-toml (9 stars, last pushed 27d ago), licensed MIT. It adds 45 tokens to every session and 2,697 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.