agentleFS
Community-AccessCommunity-Access/accessibility-agents/.github/skills/python-development/SKILL.md

python-development

Skill409 starsChanged 6 months agoMITView on GitHub

Python and wxPython development: packaging (PyInstaller/Nuitka), testing, desktop accessibility APIs, cross-platform paths, and framework patterns.

13 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use javascript-typescript, typescript-development and 8 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • pepebot-spaceSkill

    javascript-typescript

    JavaScript and TypeScript development with ES6+, Node.js, React, and modern web frameworks. Use for frontend, backend, or full-stack JavaScript/TypeScript projects.

    Also in 4 projects that use this

  • carloss765Skill

    typescript-development

    Comprehensive guidelines for writing type-safe, scalable TypeScript applications. Use when working with .ts files, TypeScript projects, or when adding type safety to JavaScript code.

    Also in 4 projects that use this

  • kinhluanSkill

    golang-development

    Implement, review, debug, or modernize Go code and modules with idiomatic APIs, concurrency, error handling, testing, profiling, and tooling. Use for Go services, libraries, CLIs, goroutine/channel issues, module management, or performance work.

    Also in 3 projects that use this

  • carloss765Skill

    rust-development

    Comprehensive guidelines for writing safe, concurrent, and performant Rust code. Use when working with .rs files, building systems software, or high-performance applications.

    Also in 3 projects that use this

  • harrissohail6Skill

    database-design

    Database schema design and SQL best practices. Use when designing tables, writing queries, or optimizing database performance.

    Also in 3 projects that use this

  • NVIDIASkill

    github

    Interact with GitHub using the gh CLI restricted to REST API only (no GraphQL). Use when the user wants to work with GitHub issues, pull requests, repos, releases, or Actions — especially in sandboxed environments where the GraphQL endpoint may be blocked. Trigger keywords - github, gh, pull request, PR, github issue, github actions, workflow, release, gh api.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 2 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    String concatenation in loops: Use "".join() or io.StringIO instead.

    line 141
  2. Very specific:

    GUI from worker thread: Always use wx.CallAfter() or wx.PostEvent().

    line 145
  3. Very specific:

    Timer not stopped: Stop timers in EVT_CLOSE handler to prevent callbacks after destruction.

    line 147
  4. Very specific:

    AUI not uninitialized: Call _mgr.UnInit() in close handler.

    line 148
  5. Very specific:

    Dialog not destroyed: Use context managers (with MyDialog(...) as dlg:) for automatic cleanup.

    line 149
  6. Distinctive:

    Wrong parent for sizer items: All controls in a sizer must have the same parent panel.

    line 150
  7. Very specific:

    Absolute positioning: Never use SetPosition() or SetSize() for layout.

    line 151
  8. Very specific:

    Every control has a meaningful name (preceding wx.StaticText for inputs, label= for buttons, SetToolTip() for image-only controls)

    line 248
  9. Distinctive:

    When audit mode is activated, agents use these structured detection rule sets:

    line 259

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Python Development Skill

Reference data for the Developer Hub, Python Specialist, and wxPython Specialist agents.

Python Version Quick Reference

VersionKey FeaturesEOL
3.10match/case, X | Y unions, ParamSpecOct 2026
3.11Exception groups, Self type, tomllib, faster CPythonOct 2027
3.12Type parameter syntax def f[T](), @override, f-string nestingOct 2028
3.13Experimental free-threaded mode, improved error messagesOct 2029
3.14async pdb.set_trace_async(), template strings (PEP 750)Oct 2030

pyproject.toml Skeleton

[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

[project]
name = "my-app"
version = "1.0.0"
requires-python = ">=3.10"
dependencies = []

[project.optional-dependencies]
dev = ["pytest>=8.0", "ruff>=0.6", "mypy>=1.11"]

[project.scripts]
myapp = "my_app.__main__:main"

[tool.pytest.ini_options]
testpaths = ["tests"]
addopts = "-ra --strict-markers"

[tool.ruff]
target-version = "py310"
line-length = 100

[tool.ruff.lint]
select = ["E", "F", "W", "I", "N", "UP", "B", "SIM", "TCH"]

[tool.mypy]
python_version = "3.10"
strict = true

PyInstaller Quick Reference

One-File Mode

exe = EXE(pyz, a.scripts, a.binaries, a.zipfiles, a.datas,
          name='MyApp', console=False, icon='icon.ico')

One-Folder Mode

exe = EXE(pyz, a.scripts, exclude_binaries=True,
          name='MyApp', console=False, icon='icon.ico')
coll = COLLECT(exe, a.binaries, a.zipfiles, a.datas, name='MyApp')

Common Hidden Imports

  • pkg_resources.extern
  • accessible_output2 (for a11y desktop apps)
  • keyring.backends (for credential storage)
  • platformdirs
  • httpx._transports / httpcore._backends
  • encodings (always needed)

wxPython Quick Reference

Sizer Cheat Sheet

SizerWhen to Use
wx.BoxSizer(wx.VERTICAL)Stack items top-to-bottom
wx.BoxSizer(wx.HORIZONTAL)Lay items left-to-right
wx.GridBagSizer(vgap, hgap)Form layouts with labels + controls
wx.FlexGridSizer(rows, cols, vgap, hgap)Even grid layouts
wx.WrapSizerFlow layout that wraps
wx.StaticBoxSizer(wx.VERTICAL, parent, "Label")Grouped controls with border

Thread-Safe GUI Updates

# From worker thread:
wx.CallAfter(self.update_status, "Done")
wx.PostEvent(self, CustomEvent(data=result))

# NEVER do this from a worker thread:
self.status_bar.SetStatusText("Done")  # CRASH or CORRUPTION

Standard IDs

IDPurpose
wx.ID_OKOK button
wx.ID_CANCELCancel button
wx.ID_SAVESave action
wx.ID_OPENOpen action
wx.ID_EXITExit / Quit
wx.ID_HELPHelp action
wx.ID_NEWNew document
wx.ID_UNDO / wx.ID_REDOUndo / Redo

Event Types

EventTrigger
wx.EVT_BUTTONButton click
wx.EVT_MENUMenu item selected
wx.EVT_CLOSEWindow close requested
wx.EVT_SIZEWindow resized
wx.EVT_TIMERTimer fired
wx.EVT_TEXTText control content changed
wx.EVT_LIST_ITEM_SELECTEDList item selected
wx.EVT_TREE_SEL_CHANGEDTree selection changed
wx.EVT_UPDATE_UIUI state update check

Common Pitfalls

Python

  • Mutable default arguments: def f(items=[]) shares the list across calls. Use None and create inside.
  • Late binding closures: lambda: x in a loop captures the variable, not the value. Use lambda x=x: x.
  • Circular imports: Move imports inside functions, use TYPE_CHECKING block, or restructure modules.
  • field() outside dataclass: field() is only valid inside @dataclass classes. Use plain type annotations elsewhere.
  • is vs ==: is checks identity, == checks equality. Use is only for None, True, False.
  • String concatenation in loops: Use "".join() or io.StringIO instead.

wxPython

  • GUI from worker thread: Always use wx.CallAfter() or wx.PostEvent().
  • Missing event.Skip(): Other handlers won't fire. Call event.Skip() unless you intentionally consume the event.
  • Timer not stopped: Stop timers in EVT_CLOSE handler to prevent callbacks after destruction.
  • AUI not uninitialized: Call _mgr.UnInit() in close handler.
  • Dialog not destroyed: Use context managers (with MyDialog(...) as dlg:) for automatic cleanup.
  • Wrong parent for sizer items: All controls in a sizer must have the same parent panel.
  • Absolute positioning: Never use SetPosition() or SetSize() for layout. Always use sizers.

Cross-Platform Paths

from platformdirs import user_config_dir, user_data_dir, user_cache_dir

config = user_config_dir("MyApp", "MyCompany")  # %APPDATA% / ~/Library/... / ~/.config/
data = user_data_dir("MyApp", "MyCompany")
cache = user_cache_dir("MyApp", "MyCompany")

Testing Quick Reference

# Run all tests
pytest

# Run specific test file
pytest tests/test_queue.py

# Run specific test
pytest tests/test_queue.py::test_submit_job -v

# With coverage
pytest --cov=mypackage --cov-report=term-missing

# Stop on first failure
pytest -x

# Show locals on failure
pytest -l

Logging Setup Template

import logging

def setup_logging(level: int = logging.INFO) -> None:
    logging.basicConfig(
        level=level,
        format="%(asctime)s %(name)s %(levelname)s %(message)s",
        datefmt="%Y-%m-%d %H:%M:%S",
    )
    # Quiet noisy libraries
    logging.getLogger("httpx").setLevel(logging.WARNING)
    logging.getLogger("httpcore").setLevel(logging.WARNING)

Desktop Accessibility Quick Reference

Platform API Summary

PlatformAPIPython BindingUse For
WindowsUI Automation (UIA)comtypes, pywinautoModern apps, NVDA/Narrator
WindowsMSAA / IAccessible2comtypes, pywinautoLegacy apps, JAWS
macOSNSAccessibilitypyobjcVoiceOver

wxPython Accessibility Essentials

# Name every control that lacks a visible label
# CORRECT: use StaticText immediately before the control in the sizer
label = wx.StaticText(panel, label="Scan progress:")
# ctrl = wx.Gauge(panel)  -- add label to sizer right before ctrl
# WRONG: SetName() is ignored by screen readers
# ctrl.SetName("Scan progress")  -- only affects FindWindowByName()

# Tab order follows sizer insertion order; override with:
ctrl2.MoveAfterInTabOrder(ctrl1)

# Keyboard shortcuts via accelerator table
accel = wx.AcceleratorTable([
    (wx.ACCEL_CTRL, ord('S'), wx.ID_SAVE),
    (wx.ACCEL_CTRL, ord('Q'), wx.ID_EXIT),
])
frame.SetAcceleratorTable(accel)

# Platform-correct button order in dialogs
sizer.Add(dialog.CreateStdDialogButtonSizer(wx.OK | wx.CANCEL))

Screen Reader Interaction Model

Screen readers expose controls as: Name + Role + Value + State

PropertywxPython SourceExample
NamePreceding wx.StaticText, label= parameter, or SetToolTip()"Scan progress"
RoleWidget type (automatic)button, text field, list
ValueWidget content"75%", "Hello world"
StateWidget flagsfocused, disabled, checked

Desktop A11y Checklist

  1. Every control has a meaningful name (preceding wx.StaticText for inputs, label= for buttons, SetToolTip() for image-only controls)
  2. Keyboard-only operation -- every action reachable via Tab/Enter/Space/arrows
  3. Focus visible -- never suppress focus indicators
  4. Tab order is logical (generally top-to-bottom, left-to-right)
  5. Color is not the sole information carrier
  6. High contrast mode supported (use system colors, not hardcoded)
  7. Dialogs use CreateStdDialogButtonSizer() for platform-correct button order
  8. AUI panes are keyboard-navigable

Structured Audit Rule Sets

When audit mode is activated, agents use these structured detection rule sets:

Rule PrefixAgentScopeCount
WX-A11Y-001..012wxpython-specialistwxPython-specific patterns (StaticText labels, AcceleratorTable, mouse-only events, dialogs)12 rules
DTK-A11Y-001..012desktop-a11y-specialistPlatform-level API patterns (Name/Role/State/Value, focus, UIA/NSAccessibility)12 rules
TST-A11Y-001..010desktop-a11y-testing-coachTest coverage gaps (automated tests, SR testing, keyboard plans, CI integration)10 rules

Rule sets don't overlap -- WX covers wxPython widget patterns, DTK covers platform APIs, TST covers testing process gaps.

Agent Routing

For deeper expertise, the skill routes to these specialists:

  • desktop-a11y-specialist -- Platform API implementation, wx.Accessible, custom widget patterns (DTK-A11Y-* audit rules)
  • desktop-a11y-testing-coach -- NVDA/JAWS/Narrator testing, Accessibility Insights, automated UIA tests (TST-A11Y-* audit rules)
  • a11y-tool-builder -- Rule engine architecture, document parsers, severity scoring, report generators

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.