From python-toolkit
This skill should be used when the user asks to "configure the python project", "set up ruff", "configure pyright", "scaffold a pixi project", "add a ruff config", or otherwise requests scaffolding or modification of ruff/pyright/pixi configuration files. Detects existing configs and creates or updates them with sensible defaults, prompting before overwriting.
How this skill is triggered — by the user, by Claude, or both
Slash command
/python-toolkit:configureThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Configure python development tools — ruff (lint/format), pyright (type-check), and pixi (project manager). Scaffold configuration files with sensible defaults when none exist; show the current state and ask before overwriting when they do.
Configure python development tools — ruff (lint/format), pyright (type-check), and pixi (project manager). Scaffold configuration files with sensible defaults when none exist; show the current state and ask before overwriting when they do.
Parse the user's request for a tool selector (ruff, pyright, pixi, or all). If no selector is provided, detect existing configs and present a menu so the user can choose.
Detection phase — check for existing config files in the current directory:
ruff.toml or [tool.ruff] section in pyproject.tomlpyrightconfig.json or [tool.pyright] section in pyproject.tomlpixi.tomlpyproject.toml (note which tool sections exist)Report what is present and what is missing.
If no selector was provided, present the detection results to the user and ask which tool to configure. Offer all to scaffold everything.
If pyproject.toml exists, prefer adding a [tool.ruff] section there. Otherwise create a standalone ruff.toml.
Use these sensible defaults:
# ruff.toml (or [tool.ruff] in pyproject.toml)
line-length = 88
target-version = "py312"
[lint]
select = ["E", "F", "I", "UP", "B"]
If a ruff config already exists, show the current settings and ask before overwriting.
Create pyrightconfig.json with sensible defaults:
{
"typeCheckingMode": "standard",
"pythonVersion": "3.12",
"reportMissingImports": true,
"reportMissingTypeStubs": false
}
If pixi.toml is present, add "venvPath": "." and "venv": ".pixi/envs/default" to the pyright config.
If .venv/ is present, add "venvPath": "." and "venv": ".venv" to the pyright config.
If a pyright config already exists, show current settings and ask before overwriting.
If pixi.toml does not exist, scaffold a minimal one (substitute the actual directory name for <directory-name>):
[project]
name = "<directory-name>"
version = "0.1.0"
channels = ["conda-forge"]
platforms = ["linux-64", "win-64", "osx-arm64"]
[dependencies]
python = ">=3.12"
If pixi.toml already exists, show the current config and offer to add common features (dev dependencies, tasks, etc.).
Run all three configurations in sequence (ruff → pyright → pixi), respecting existing configs (prompt before any overwrite).
Summarize what was created and what was modified.
This skill currently inlines its procedure (direct file detection + scaffolding). The forward-looking design per the v3 toolkit spec and the Path B decision is for this skill to invoke the shared diagnostic-review pipeline via:
Skill(skill: "python-toolkit:diagnostic-review", pass: "configure")
The shared diagnostic-review skill already ships at skills/diagnostic-review/SKILL.md and documents the pass: configure selector (auto-detects the project tier, diffs proposed updates against .toolkit.local.md, and prompts per-section). The per-pass invocation wiring lands in a separate chore — until then, this skill body matches the legacy /configure command behavior.
npx claudepluginhub zaynram/code-marketplace --plugin python-toolkitGuides collaborative design exploration before implementation: explores context, asks clarifying questions, proposes approaches, and writes a design doc for user approval.
Creates structured, bite-sized implementation plans from specs or requirements before writing code. Useful for breaking down multi-step tasks into testable steps with file structure and task boundaries.
Synthesizes the current conversation into a structured spec (PRD) and publishes it to the project issue tracker with a ready-for-agent label, without interviewing the user.