azure-sdk-python-sample-review

name	azure-sdk-python-sample-review
description	Reviews Azure SDK Python code samples for best practices, credential handling, async patterns, error handling, and documentation compliance. Trigger: "review Python Azure SDK sample", "check Python sample", "Azure SDK Python review".
status	active
tags	["review","python","azure","sdk","sample","check","best","practices","code","reviewing"]
category	review

USE FOR

• "review Python Azure SDK sample"
• "check Python sample for best practices"
• "Azure SDK Python code review"
• Reviewing credential handling in Azure SDK Python samples
• Error pattern and async pattern analysis for Python Azure SDK samples
• Documentation compliance check for Python Azure SDK samples

DO NOT USE FOR

• General Python code review unrelated to Azure SDK samples
• Production application code review
• Azure service configuration

Context

Base template: Inherits from azure-sdk-sample-review for shared review patterns (credentials, error handling, documentation, infrastructure). This skill adds Python-specific rules below.

Reviews Python code samples for Azure SDKsintended for publication as Microsoft Azure samples. Focuses on Track 2 azure-* packages, DefaultAzureCredential, service-specific patterns (Cosmos DB, SQL, Storage, Service Bus, Key Vault, AI), sample hygiene, documentation accuracy, IaC (Bicep/Terraform), azd integration, and async patterns.

Total rules: 75 (11 CRITICAL, 23 HIGH, 32 MEDIUM, 9 LOW)

Severity Legend

• CRITICAL = security/won't run
• HIGH = confuses users
• MEDIUM = best practice
• LOW = polish

---

Quick Pre-Review Checklist (5-Minute Scan)

• Track 2 Azure SDK packages (azure-*)
• DefaultAzureCredential (not connection strings/keys)
• .gitignore includes .env, __pycache__/, .venv/
• No hardcoded credentials in code
• README.md with prerequisites, setup, expected output
• MIT LICENSE file present
• pip-audit passes (no critical/high CVEs)
• Type hints on public functions
• try/except with specific exception types
• Clients closed (with/async with)
• Pinned requirements.txt or lock file committed
• Single package manager (pip OR poetry OR uv)
• Python 3.10+ required, 3.12/3.13 target

---

Blocker Issues (Auto-Reject)

1. Hardcoded secrets in code
2. Missing authentication or insecure auth methods
3. No error handling / bare except: blocks
4. Broken imports / missing dependencies
5. Critical/high CVEs in pip-audit
6. Missing LICENSE at any repo hierarchy level. ⚠️ Check repo root first.
7. .env committed to git. ⚠️ Verify with git ls-files .env.
8. Track 1 packages (legacy pre-Track-2 APIs)

---

Detailed Rules

Language-Specific References (Python code examples)

#	Section	Reference File
1	Project Setup — Python	references/project-setup.md
2	SDK Client Patterns — Python	references/sdk-client-patterns.md
3	AI Services — Python	references/ai-services.md
4	Data Services — Python	references/data-services.md
5	Messaging — Python	references/messaging.md
6	Key Vault — Python	references/keyvault.md
7	Vector Search — Python	references/vector-search.md
8	Error Handling — Python	references/error-handling.md
9	Data Management — Python	references/data-management.md
10	Sample Hygiene — Python	references/sample-hygiene.md
11	Documentation — Python	references/documentation.md
12	Infrastructure — Python	references/infrastructure.md
13	azd — Python	references/azd.md
14	Async Patterns (Python-unique)	references/async-patterns.md
15	CI/CD & Testing — Python	references/cicd-testing.md

---

Companion Skills

• azure-sdk-typescript-sample-review — TypeScript equivalent (shared IaC/azd patterns)
• azure-sdk-dotnet-sample-review — .NET 9/10 + Aspire
• azure-sdk-java-sample-review — Java 17/21 + Spring Boot
• azure-sdk-go-sample-review — Go 1.21+
• acrolinx-score-improvement — Article quality and style

---

Summary

75 rules covering Azure SDK Python sample quality: authentication, client patterns, service-specific best practices (AI, Data, Messaging, Key Vault, Vector Search), infrastructure-as-code, async patterns, documentation, and CI/CD. Apply to ensure samples are secure, accurate, and publication-ready.

References

References location: All reference files for this skill live inside the skill directory at .github/skills/data-plus-ai-sdk-python-sample-review/. Paths like references/file.md resolve to .github/skills/data-plus-ai-sdk-python-sample-review/references/file.md. Paths are relative to the skill folder, not the repo root.