Skip to content

test-plan

End-to-end test planning and automation workflow for RHOAI (Red Hat OpenShift AI). Takes a Jira strategy document as input and produces a complete test plan, individual test case specifications, executable test automation code, and visual UI verification reports. The pipeline uses parallel sub-agent analysis (endpoints, risks, infrastructure) to extract testable interfaces, then scores quality with a 5-criteria rubric and auto-revises up to 2 cycles. Supports the full lifecycle: create, review, publish to GitHub, resolve PR feedback, update with new documentation, and verify UI test cases against live clusters via Playwright. Test code generation uses odh-test-context for repo-specific conventions, Tiger Team pattern guides, and intelligent placement analysis to decide whether tests belong in the component repo or downstream E2E repo.

Plugin Details

Pipeline

test-plan pipeline

Skills

Skill Description Invocable
/test-plan-create Generate a test plan from a strategy
/test-plan-create-cases Generate test case files from a test plan
/test-plan-update Update test plan with new docs (ADR, API specs), re-analyze, bump version
/test-plan-case-implement Generate executable test automation code from TC specifications with intelligent placement
/test-plan-ui-verify Verify UI test cases from a PR against a live ODH/RHOAI cluster via Playwright; supports upgrade testing workflow
/test-plan-publish Publish test plan artifacts to GitHub with PR creation
/test-plan-resolve-feedback Assess and resolve PR review comments on test plans
/test-plan-score Score test plan quality using rubric without auto-revision
/test-plan-analyze-endpoints Extract scope and API endpoints internal
/test-plan-analyze-risks Determine test levels, priorities, NFRs, and risks internal
/test-plan-analyze-infra Identify environment and infrastructure needs internal
/test-plan-merge Intelligently merge new analyzer findings into existing test plan internal
/test-plan-resolve-gaps Cross-reference gaps with new findings to determine what's resolved internal
/test-plan-analyze-placement Analyze test cases and recommend placement (component repo vs downstream) internal
/test-plan-review Review test plan with 5-criteria rubric and auto-revision internal
/test-plan-generate-test-file Generate complete test file with all functions, quality scoring and auto-revision internal
/test-plan-score-test-function Score generated test function code using 5-criteria quality rubric internal

Installation

/plugin install test-plan@opendatahub-skills

Architecture

The plugin is organized as an orchestrated pipeline of 19 skills, split into user-invocable commands and internal forked sub-agents:

User-invocable pipeline stages: 1. test-plan-create -- generates a test plan from a Jira strategy with 3 parallel analyzers 2. test-plan-create-cases -- generates TC-*.md files from the test plan 3. test-plan-update -- re-analyzes with new documents, merges findings, bumps version 4. test-plan-case-implement -- generates executable test code with placement analysis 5. test-plan-ui-verify -- browser-based UI test execution via Playwright CDP 6. test-plan-publish -- creates/updates GitHub PRs with test plan artifacts 7. test-plan-resolve-feedback -- triages and applies PR review comments 8. test-plan-score -- standalone quality rubric assessment (read-only)

Internal sub-agents (context: fork): - test-plan-analyze-endpoints, test-plan-analyze-risks, test-plan-analyze-infra -- parallel analyzers - test-plan-merge -- intelligent merge of new findings into existing plans - test-plan-resolve-gaps -- cross-references gaps with new documentation - test-plan-analyze-placement -- recommends component vs downstream repo placement - test-plan-review -- quality scoring + auto-revision loop (max 2 cycles) - test-plan-generate-test-file -- parallel test file generation sub-agent - test-plan-score-test-function -- quality scoring for generated test functions

Key architectural patterns: - Parallel forked sub-agents for analysis and code generation (isolated context, clean return) - Python scripts for deterministic operations (frontmatter, validation, file mapping, repo utilities) - MCP integration (Atlassian) for fetching Jira strategies - Persistent Playwright CDP browser for UI verification with screenshot capture - odh-test-context integration for repository-specific test conventions