More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    python-style: Python 3.15 Coding Standard

    1

    A coding standard for free-threaded CPython 3.15 that your agent follows whenever it writes, reviews or upgrades Python.

    Free

    Security scanned
    python-style: Python 3.15 Coding Standard

    python-style: Python 3.15 Coding Standard

    Example session with this skill installed

    Review this helper module. We just moved to 3.15t.

    _MISSING = object()
    
    def is_config(value):
        return isinstance(value, dict)
    
    def merge_all(groups):
        return sum(groups, [])
    
    def get(cfg, key, default=_MISSING):
        if key in cfg:
            return cfg[key]
        if default is _MISSING:
            raise KeyError(key)
        return default
    
    • Read your context and instructions
    • Compiled the python-style

    Findings

    SeverityRuleLocationFindingFix
    StalePS-DATA-12helpers.py:4isinstance(value, dict) is False for a frozendict, so an immutable config is rejectedTest isinstance(value, collections.abc.Mapping)
    StalePS-TYPE-20helpers.py:1Sentinel built with object(): the type checker cannot name it, so default ends up typed AnyMISSING = sentinel("MISSING") at module scope, and annotate the parameter with it
    PerformancePS-PERF-06helpers.py:7sum(groups, []) builds the result by repeated concatenation, which is quadratic[*group for group in groups]

    Discussion

    The first row is a correctness bug that only appears once someone passes a frozendict, which 3.15 makes the natural type for configuration. It raises no error: is_config just answers False, and the caller takes the wrong branch. The other two are worth doing in the same change, since the sentinel fix is what lets ty check every caller of get.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    A coding standard for free-threaded CPython 3.15 that your agent follows whenever it writes, reviews or upgrades Python.

    It describes no shims for older versions. Most of what 3.14 and 3.15 changed alters behaviour without raising an error: open() is UTF-8 whatever the locale, isinstance(x, dict) is False for a frozendict, a contextmanager-built decorator holds until its generator is exhausted. Nothing in your test suite will flag these on its own, so section 2 of the skill lists them up front.

    It is built on the normative specs (PEP 8, PEP 257, PEP 20 and the Python Typing Spec, cited from the living specs), the conventions of the CPython 3.15 standard library, and Ruff's rule catalogue as the house standard. Where they disagree, it picks a side and says why.

    What's inside

    • 262 rules across 14 chapters: naming, types, data classes and enums, errors, concurrency without the GIL, asyncio, output and logging, performance, public API, modules and lazy import, documentation, testing with pytest 9, build and packaging, and migration.
    • Covers what 3.15 added: frozendict, sentinel(), TypeForm, closed TypedDict, lazy import, unpacking in comprehensions, the profiling package and TaskGroup.cancel().
    • Bundled assets: ruff.toml, ty.toml (every opt-in rule named one by one, enabled or left off with a reason), a pyproject.toml with PEP 639 licensing and PEP 735 dependency groups, and a ci.yml that installs the free-threaded interpreter and asserts the GIL is really off.

    How it works

    • Section 2 — reality check. A table of every spelling the target release changed: what still compiles but now behaves differently, what no longer compiles, and what was gained. Your agent reads it before writing a single line, because most code on the public internet (and most of what a model has read) is stale.
    • Section 4 — core rules. The rules that bind on their own and reach nearly every file.
    • One chapter at a time. Each topic chapter opens with a table of contents of the rules it owns. The agent loads only the chapter the task needs, so the standard costs context in proportion to the task, not its size.
    • Tiered rules. [M] mandatory, [M!] mandatory in library and high-reliability code (each states a relaxed form for scripts and prototypes), [S] default that needs a one-line justification to break.
    • Every rule has a Why and a Check. The Check is something concrete to grep for or look at, so a review is reproducible rather than taste.
    • Labelled examples. Each code block says whether it was verified against the target toolchain, is a fragment, or is shown because a tool rejects it.
    • Review protocol. A seven-level severity model (Unsound, Stale, Incorrect, Contract, Design, Performance, Style), a fixed review order, a mechanical grep pass, and an output table of severity, rule ID, location, finding and fix. Includes the order in which to refactor an existing codebase.

    Where it stops

    • It is a standard your agent follows, not a linter or formatter. It tells the agent which tools to run and how to configure them, but it runs nothing on its own.
    • It targets one release. Older releases appear only in the migration chapter, as spellings to move away from.
    • CPython 3.15 only. Nothing below 3.15 is supported, and no compatibility shim is described.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.

    Targets free-threaded CPython 3.15 only; no guidance for older interpreters. It is a standard the agent follows, not a linter, and does not run tools on its own. The bundled configs pin tool versions you should update as new releases land.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 2 days ago

    • Free with an Agensi account

    Software
    Python
    Optional
    CPython 3.15 (free-threaded)uv, ruff, tyripgrep (rg)

    Permissions

    Read Files
    Write Files
    Terminal / Shell

    File Scopes

    **/*.py
    pyproject.toml
    ruff.toml
    ty.toml
    .github/workflows/*.yml

    Instructions and reference text only: no scripts, no network access, no credentials. Read and write cover the code you ask it to write, review or refactor. The shell is optional, for running the linter, type checker, tests and the rg checks the rules name. Bundled config templates are copied into a project only when you ask.

    Listed2 days ago

    What's inside

    Frequently Asked Questions

    No. It targets CPython 3.15 on the free-threaded build and deliberately describes no compatibility shims.

    Ruff for linting and formatting, and ty for type checking. Both configurations are bundled in assets/.

    The rules assume it, because code that is correct there is also correct with the GIL. The bundled CI installs the free-threaded interpreter and checks at runtime that the GIL is off.

    No. The agent reads SKILL.md, then loads only the chapter that matches the task.

    New in Code Quality & Review

    Free