git:20260725.11dd3e7 to git:20260807.4fb0cdd

1 added, 1 removed. Audit A to A.

# Katalon True Skills

- > Open, agent-neutral AI testing skills for Katalon True Platform (TestOps / TrueTest). Thirteen skills cover the full software testing lifecycle - test plan, test design, test management and requirement traceability, test review, test execution, failure analysis and release readiness, and test maintenance - operated from any AI coding agent's chat through the Katalon MCP server. Author the skill bodies once in `skills/`; a deterministic build generates native config for Claude Code, Codex, GitHub Copilot, Cursor, Kiro, Windsurf, Cline, Continue, and any agent that reads AGENTS.md. Works on any model tier, from the largest models down to small ones, because every skill uses deterministic tool order, explicit stop conditions, and stated capability boundaries.

+ > Agent-neutral AI testing skills for Katalon True Platform (TestOps / TrueTest). Thirteen skills cover the full software testing lifecycle - test plan, test design, test management and requirement traceability, test review, test execution, failure analysis and release readiness, and test maintenance - operated from any AI coding agent's chat through the Katalon MCP server. Author the skill bodies once in `skills/`; a deterministic build generates native config for Claude Code, Codex, GitHub Copilot, Cursor, Kiro, Windsurf, Cline, Continue, and any agent that reads AGENTS.md. Works on any model tier, from the largest models down to small ones, because every skill uses deterministic tool order, explicit stop conditions, and stated capability boundaries.

## Start here

- [README](https://github.com/katalon-labs/true-skills/blob/main/README.md): what it is, the 7-stage lifecycle diagram, the skills table, install per agent, and the Katalon MCP setup.

- [AGENTS.md](https://github.com/katalon-labs/true-skills/blob/main/AGENTS.md): the machine-readable skill index for any agent - match a request to a skill description, then open that skill's SKILL.md.

- [Lifecycle surface](https://github.com/katalon-labs/true-skills/blob/main/docs/lifecycle.html): human-facing visual of the 7 stages, the skills that own each, and the MCP tools behind them.

## Skills (single source of truth: skills/)

- [platform-setup](https://github.com/katalon-labs/true-skills/blob/main/skills/platform-setup/SKILL.md): install, connect, and verify the Katalon MCP; diagnose auth/access. Start here.

- [test-plan](https://github.com/katalon-labs/true-skills/blob/main/skills/test-plan/SKILL.md): stage 1 - scope, risk-prioritized planning, and the executable folder+suite that stands in for a Test Plan.

- [create-test-cases](https://github.com/katalon-labs/true-skills/blob/main/skills/create-test-cases/SKILL.md): stage 2 - design atomic manual test cases from requirements (ISTQB as reference), avoid duplicates, import and link.

- [test-management](https://github.com/katalon-labs/true-skills/blob/main/skills/test-management/SKILL.md): stage 3 - organize inventory and produce a requirement-to-test traceability report with coverage and orphans.

- [test-review](https://github.com/katalon-labs/true-skills/blob/main/skills/test-review/SKILL.md): stage 4 - pre-pipeline coverage/quality/flakiness review returning an Approve / Approve-with-fixes / Reject verdict.

- [execute-test](https://github.com/katalon-labs/true-skills/blob/main/skills/execute-test/SKILL.md): stage 5 - manual runs, Run with AI, and scheduled automation; summarize pass/fail/blocked.

- [upload-report](https://github.com/katalon-labs/true-skills/blob/main/skills/upload-report/SKILL.md): stage 5/6 - run automation and upload/verify Katalon, JUnit, or Playwright reports.

- [test-case-to-playwright](https://github.com/katalon-labs/true-skills/blob/main/skills/test-case-to-playwright/SKILL.md): convert Katalon manual cases into Playwright TypeScript (Page Object Model + fixtures).

- [playwright-execute](https://github.com/katalon-labs/true-skills/blob/main/skills/playwright-execute/SKILL.md): run Playwright specs, upload to Katalon with @katalon/playwright-reporter, and verify the run.

- [analyze-failures](https://github.com/katalon-labs/true-skills/blob/main/skills/analyze-failures/SKILL.md): stage 6 - triage failures (product vs automation vs environment), cluster, and file ALM defects.

- [release-analyze](https://github.com/katalon-labs/true-skills/blob/main/skills/release-analyze/SKILL.md): stage 6 - produce a Ready / Ready with risk / Not ready release recommendation from quality metrics.

- [test-maintenance](https://github.com/katalon-labs/true-skills/blob/main/skills/test-maintenance/SKILL.md): stage 7 - detect flaky/broken cases, repair or regenerate, and feed gaps back into planning.

- [true-platform-testing](https://github.com/katalon-labs/true-skills/blob/main/skills/true-platform-testing/SKILL.md): the lifecycle router + end-to-end runner across all 7 stages.

## Combination recipes and prompts

- [combination-recipes](https://github.com/katalon-labs/true-skills/blob/main/skills/true-platform-testing/references/combination-recipes.md): multi-skill playbooks (requirement-to-ship, coverage rescue, flaky cleanup, manual-to-automation, cross-lane trust check, traceability audit).

- [prompt-recipes](https://github.com/katalon-labs/true-skills/blob/main/skills/true-platform-testing/references/prompt-recipes.md): copy-paste prompts plus cross-model and cross-agent execution notes.

- [mcp-tool-index](https://github.com/katalon-labs/true-skills/blob/main/skills/true-platform-testing/references/mcp-tool-index.md): every Katalon MCP tool, one line, tagged by lifecycle stage.

## Install

- Any agent: `npx skills add katalon-labs/true-skills`

- Claude Code: `claude plugin marketplace add katalon-labs/true-skills` then install `katalon-true-platform`

- Codex: `codex plugin marketplace add katalon-labs/true-skills`

- Copilot / Cursor / Kiro / Windsurf / Cline / Continue: copy the generated config directory; see README.

- MCP: `npx -y mcp-remote https://<your.sub.domain>.katalon.io/mcp --transport http-first` (browser OAuth).