Python Implementation

Provides guidance for modern Python development with a focus on correctness, performance, and maintainability.

Use this skill when

• Building or refactoring Python 3.x services, CLIs, or libraries
• Designing async workflows or concurrency patterns
• Setting up Python tooling, linting, or testing workflows

Activation cues

• "Implement this in Python"
• "Add async/await" or "use asyncio"
• "Set up pytest/ruff/mypy"
• "Build a FastAPI/Django endpoint"

Do not use this skill when

• The task is primarily about another runtime or language
• The request is strictly about data science notebooks or ML training pipelines

Inputs to confirm

• Target Python version and runtime constraints
• Packaging expectations (pyproject.toml, uv, pip, or poetry)
• Dependency constraints (allowed third-party packages or stdlib-only)
• Execution model (sync vs async, web framework or standalone)
• Quality gates (tests, lint, type checking)

Workflow (Deterministic)

1. Clarify scope and constraints.
   • Output: confirmed runtime, packaging, and acceptance criteria.
2. Choose architecture and libraries.
   • Decision: If the project already mandates a framework, follow it; otherwise keep dependencies minimal and prefer the standard library when possible.
   • Decision: If third-party dependencies are restricted, keep to stdlib and lightweight patterns.
   • Decision: If a web API is needed, pick FastAPI or Django; if CLI/script, use standard library + argparse or approved CLI helpers.
   • Output: selected stack and module boundaries.
3. Choose concurrency model.
   • Decision: If I/O-bound or high concurrency, use async/await; otherwise stay synchronous.
   • Output: sync/async approach and key primitives.
4. Implement with types and validation.
   • Output: typed interfaces, input validation at boundaries, and clear error handling.
5. Add tests and tooling guidance.
   • Output: test plan or tests added plus lint/type-check notes.
6. Summarize changes and verification.
   • Output: concise change summary and commands (if any) to validate.

Common pitfalls to avoid

• Mixing sync and async without clear boundaries
• Skipping input validation at API/CLI boundaries
• Adding heavy dependencies when the stdlib is enough
• Ignoring project dependency constraints
• Providing unbounded concurrency or missing timeouts
• Assuming dependencies or tooling are already available

Examples

Example input: "Add a FastAPI endpoint that accepts JSON and validates fields."

Expected output: A FastAPI route with Pydantic models, validation errors handled, and a note on running pytest and ruff if available.

Example input: "Convert this blocking HTTP loop to async."

Expected output: Use httpx.AsyncClient if approved, async entrypoint, and note any required event loop changes.

Output Contract (Always)

Report using this format:

• Summary: 1–3 bullets describing what changed
• Files changed: list of paths updated or created
• Tests/verification: commands run or suggested (state if not run)
• Follow-ups: questions or next steps if anything is blocked

References (Optional)

• Reference index: references/README.md