Osdu QA

OSDU QA

Route based on the first argument:

Arguments received: $ARGUMENTS

Getting Started

New to OSDU QA? See the Workflow Guide for the complete testing process from setup through report generation and PowerPoint presentations.

Quick workflow: Environment Setup → Health Check → Smoke Tests → Gate Tests → Report Generation

Quick Reference

Environment Commands

/osdu-qa env                    # List environments
/osdu-qa env use <platform>/<env> # Switch environment
/osdu-qa env status             # Show current config
/osdu-qa env audit              # Audit all environments (list format)
/osdu-qa env audit --check      # Audit with API health checks

Test Commands

/osdu-qa test                   # Show test plan
/osdu-qa test smoke             # Run smoke tests
/osdu-qa test legal             # Run legal API tests
/osdu-qa test versions          # Show service versions
/osdu-qa test compare qa temp   # Compare environments

Quick Commands

/osdu-qa check                  # Connectivity check
/osdu-qa list                   # List all test collections
/osdu-qa report                 # Generate QA report

CLI Scripts

All commands map to Python scripts:

# Environment management
uv run skills/osdu-qa/scripts/env_manager.py <command>

# Test execution
uv run skills/osdu-qa/scripts/osdu_test.py <command>

# Service versions
uv run skills/osdu-qa/scripts/service_versions.py

# Report generation
uv run skills/osdu-qa/scripts/generate_report.py --format both

Specialized Agents

IMPORTANT: Always delegate to agents for test execution and analysis. Direct CLI is only for debugging.

How to Delegate

Use the Agent tool to spawn a sub-agent. Never simulate or inline a sub-agent's work.

Agent(
  subagent_type="qa-runner",
  description="Run smoke tests on cimpl/qa",
  prompt="Run the smoke tests on cimpl/qa and provide a structured summary with pass rates and any failures"
)

Replace subagent_type with the appropriate agent (qa-runner, qa-analyzer, qa-comparator, or qa-reporter) and adjust the prompt to match the request. The sub-agent runs independently and returns a structured summary.

Examples

# Single test collection - USE AGENT
Use qa-runner to run the smoke tests

# Multiple collections - USE AGENT
Use qa-runner to execute all P0 collections in parallel

# Failure investigation - USE AGENT
Use qa-analyzer to investigate the Storage test failures

# Reports - USE AGENT
Use qa-reporter to generate an HTML dashboard

When to Use Direct CLI

Only use direct CLI (osdu_test.py) for:

• Debugging script issues
• Quick connectivity checks (check command)
• Listing available tests (list command)

Environments

Environments are user-local configuration stored in config/environments.json (gitignored).

Run /osdu-qa env to see configured environments. Add with: /osdu-qa env add <platform>/<name> --host <host> --partition <partition> --auth-type <azure-ad|keycloak> Remove with: /osdu-qa env remove <platform>/<name> --confirm

See reference/environments.example.json for the configuration schema.

Test Collections (30 aliases, 33 collection files, ~2,295 tests)

Smoke - Pre-flight (1 collection, 153 tests)

P0 - Core Platform (7 collections, 745 tests)

P1 - Core+ (9 collections, 958 tests)

P2 - DDMS & Domain (13 collections, 608 tests)

Group Aliases

Supporting Files

• env.md - Environment management details
• test.md - Test execution details
• reference/workflow-guide.md - Complete QA workflow process
• reference/collections.md - Full collection catalog
• reference/troubleshooting.md - Common issues

File Structure

skills/osdu-qa/
├── SKILL.md              # This file (routing)
├── env.md                # Environment management instructions
├── test.md               # Test execution instructions
├── scripts/              # Python CLI tools
├── config/               # Runtime configuration
├── templates/            # Report templates
└── reference/            # Documentation
    ├── workflow-guide.md # Complete QA workflow process
    ├── collections.md    # Full collection catalog
    └── troubleshooting.md # Common issues

agents/                   # Specialized agents
├── qa-runner.agent.md
├── qa-analyzer.agent.md
├── qa-comparator.agent.md
└── qa-reporter.agent.md