qa-engineer 0.9.1 → 0.10.0

package/COMPATIBILITY.md

@@ -6,7 +6,15 @@ This document records which AI coding agents the pack targets, how each one disc
 
 ## Release compatibility
 
-No versions have been released yet. Beginning with the first tagged release, this section will carry a matrix of pack versions against tested agent versions and the Agent Skills specification revision each release was validated against.
+| Pack version | Released | Discovery paths verified against agent docs | Live agent runs |
+| --- | --- | --- | --- |
+| 0.9.0, 0.9.1 | 2026-07-28 | Not systematically re-read for the release | Claude Code, Cursor (manual, by the maintainer) |
+| 0.9.2 | 2026-07-28 | Yes — every host's own documentation, re-read on 2026-07-28 (sources in the matrix below) | **Do not use.** The tarball omitted the engine; every command failed on a missing module |
+| 0.9.3 | 2026-07-28 | Yes, as 0.9.2 | Claude Code, Cursor (manual, by the maintainer) |
+
+"Verified" in the third column means the discovery paths in the matrix below were read off each host's own documentation on that date and are asserted against the installer by `packages/installer/test/hosts.test.mjs`, which fails if the installer writes to a path that host does not document reading. It does **not** mean the pack was executed inside every host: that is the Tier 1 bar, and only Claude Code and Cursor have had a real run.
+
+Documentation is re-read at each release. A test can pin the installer to a recorded claim; it cannot notice that a vendor moved a directory, so that check is a release step performed by a person.
 
 ## The standard this pack targets
 
@@ -14,23 +22,26 @@ Skills are authored against the open [Agent Skills specification](https://agents
 
 ## Target agent matrix
 
-Verified against official documentation as of July 2026. Entries marked *unverified* rely on secondary sources and will be confirmed before the relevant integration ships.
+Each row's discovery paths were read off that host's own documentation on **2026-07-28** and are recorded in the installer's agent registry, where a test asserts the installer writes to one of them. The **bold** path is the one `qa install` targets.
+
+| Agent | Paths this host reads (project) | Pack installs to | How you reach a skill | Source | Planned tier |
+| --- | --- | --- | --- | --- | --- |
+| Claude Code | `.claude/skills/` | **`.claude/skills/`** | `/qa-explore`, or describe the task | [docs](https://code.claude.com/docs/en/skills) | Tier 1 |
+| OpenAI Codex CLI | `.agents/skills/` — at the working directory, its parent, and the repo root. There is **no** `.codex/skills/` | **`.agents/skills/`** | `$qa-explore`, or `/skills` to browse | [docs](https://developers.openai.com/codex/skills) | Tier 1 |
+| OpenCode | `.agents/skills/`, `.opencode/skills/`, `.claude/skills/` | **`.agents/skills/`** | `/qa-explore` (generated command), or the agent's own `skill` tool | [docs](https://opencode.ai/docs/skills/) | Tier 1 |
+| Cursor | `.agents/skills/`, `.cursor/skills/`, plus Claude and Codex directories | **`.agents/skills/`** | `/` in Agent chat, then pick the skill | [docs](https://cursor.com/docs/skills) | Tier 2 |
+| Antigravity | `.agents/skills/` (default), `.agent/skills/` (legacy) | **`.agents/skills/`** | describe the task — skills auto-activate | [docs](https://antigravity.google/docs/skills) | Tier 2 |
+| GitHub Copilot | `.agents/skills/`, `.github/skills/` | **`.agents/skills/`** | `/qa-explore` in Copilot Chat (generated prompt file) | [docs](https://docs.github.com/en/copilot) | Tier 2 |
+| Gemini CLI | `.agents/skills/`, `.gemini/skills/` | **`.agents/skills/`** | `/qa-explore` (generated command) | [docs](https://google-gemini.github.io/gemini-cli/) | Tier 2 |
+| Kimi (Agent Skills copy) / other spec-compliant hosts | `.agents/skills/` | **`.agents/skills/`** | product's own surface | [spec](https://agentskills.io/specification) | Tier 2 |
 
-| Agent | Native `SKILL.md` support | Skill discovery path the pack targets | Invocation surface | Planned tier |
-| --- | --- | --- | --- | --- |
-| Claude Code | Yes (format originator) | `.claude/skills/` (project), plugins | `/skill-name`, auto-activation | Tier 1 |
-| OpenAI Codex CLI | Yes | `.agents/skills/` | `$skill-name` mention, `/skills` menu, auto-activation | Tier 1 |
-| OpenCode | Yes | `.agents/skills/`, `.opencode/skills/` | native `skill` tool (auto); slash via generated commands | Tier 1 |
-| Cursor | Yes (2.4+) | `.agents/skills/`, `.cursor/skills/` | `/skill-name`, auto-activation | Tier 2 |
-| GitHub Copilot | Yes (VS Code, JetBrains, CLI, cloud) | `.agents/skills/`, `.github/skills/` | `/skill-name` in VS Code; auto-activation elsewhere | Tier 2 |
-| Gemini CLI | Yes | `.agents/skills/`, `.gemini/skills/` | auto-activation; slash via generated commands | Tier 2 |
-| Antigravity | Yes | `.agents/skills/` | auto-activation; slash via workflows (*path unverified*) | Tier 2 |
-| Kimi / other Agent Skills hosts | Yes (copy) | `.agents/skills/` (or product-specific path) | auto-activation / product slash | Tier 2 |
+Three facts follow from that table and shape the design:
 
-Two facts shape the integration design:
+1. **One path serves seven of the eight hosts.** Everything except Claude Code reads `.agents/skills/`, so `qa install` writes the same canonical skill set there — copies, never conversions. Claude Code gets `.claude/skills/` as well, and a project where only Claude Code is detected still receives the shared path, so opening the same repository in Cursor or Codex tomorrow needs no reinstall.
+2. **Auto-activation is the only channel every host shares.** Codex takes `$qa-explore`, Cursor matches on `/`, OpenCode's agent calls a `skill` tool. `qa install` prints the convention for the host it detected, because a user who types the wrong one has no way to tell a wrong keystroke from a failed install.
+3. **Slash ergonomics on Gemini CLI, OpenCode, and Copilot need a thin generated wrapper.** Those wrappers are rendered from skill frontmatter alone (ADR-0002), capped at 15 lines, and carry no knowledge. Antigravity's workflow directory (`.agents/workflows/`) is *not* documented by Google, so those wrappers are generated only on explicit `--agent antigravity`; skills themselves work there without them.
 
-1. Six of the seven primary agents read `.agents/skills/` directly; Claude Code additionally requires `.claude/skills/`. A single canonical skill set therefore covers every target with copies, not conversions. Additional hosts (Kimi and peers) use the same copy path — see [docs/installation/other-agents.md](docs/installation/other-agents.md).
-2. Description-driven auto-activation is the only invocation channel available on all primary agents. Slash-command ergonomics on Gemini CLI, OpenCode, and Antigravity require thin generated wrappers from the installer (`qa install`).
+**Antigravity detection is deliberately conservative.** Its skills live at `.agents/skills/`, which is also the path this installer creates for every host — so detecting Antigravity from `.agents/` would mean detecting it because the pack had run once before. An earlier release did exactly that and reported a host that was never there. Skills work in Antigravity either way; only the optional workflow wrappers need `--agent antigravity`.
 
 ## Support tiers
 
@@ -45,8 +56,14 @@ The tier column in the matrix above is the **planned** tier — the level each a
 
 | Requirement | Status | Used for |
 | --- | --- | --- |
-| Python 3.8+ | Required for analysis | The deterministic analysis toolkit (`shared/analysis/lib/`) and framework analyzers — standard library only, no packages |
-| Node.js 18+ | Required for the installer CLI | `npx qa-engineer` / `qa install` / `qa doctor` / `qa self-test` (interactive onboarding uses `--yes` / `--ci` in non-TTY and CI) |
+| Node.js 18.17+ | The only requirement | Everything: `npx qa-engineer` and the CLI, and the deterministic engine the skills run (`packages/engine/`) |
+
+**One runtime, no dependencies.** Until 0.9.2 the engine was Python, which meant a
+second runtime users had to install for tooling they had installed with Node — and
+when it was missing, skills fell back to model guesswork and said nothing.
+[ADR-0012](docs/architecture/ADR-0012-node-engine.md) records why that changed and
+how the migration was verified. The engine still takes no third-party dependencies:
+the XML and ZIP readers it needs are written out rather than depended on.
 
 ## Operating systems
 
@@ -56,15 +73,22 @@ The tier column in the matrix above is the **planned** tier — the level each a
 | macOS | Expected to work; not covered by CI |
 | Windows | Expected to work; **not yet verified end to end** |
 
-The installer uses Node's platform-independent path handling, and the skills
-invoke their bundled engine through a launcher (`scripts/qa_tool.py`) that needs
-no shell features — the command shape is identical in bash, zsh, PowerShell, and
-cmd.exe. The one difference on Windows is the interpreter name: use `python` where
-`python3` is not on PATH.
-
-Until a Windows run is confirmed, that row says "expected", not "supported".
-
-The analysis toolkit is standard-library-only Python, so it runs on any Python 3.8+ interpreter with nothing to install. Requirements are declared per skill via the specification's `compatibility` frontmatter field, and the pack degrades gracefully: skills describe a manual fallback whenever a runtime or optional integration (such as an MCP server) is unavailable.
+The installer uses Node's platform-independent path handling, and the skills invoke
+the engine through a committed launcher (`scripts/qa-tool.mjs`) that needs no shell
+features — the command shape is identical in bash, zsh, PowerShell, and cmd.exe,
+with no platform difference at all.
+
+The Windows row said "expected" for a specific reason that no longer applies: the
+engine was Python, and `python3` is not on PATH on Windows by default, so every
+deterministic call there failed and each skill quietly fell back to guesswork. The
+engine is now the same Node that ran the install. The row still says "expected"
+rather than "supported" because nobody has run the pack end to end on Windows —
+that is an unverified claim, not a known gap.
+
+Requirements are declared per skill via the specification's `compatibility`
+frontmatter field, and the pack degrades gracefully: skills describe a manual
+fallback whenever a runtime or optional integration (such as an MCP server) is
+unavailable.
 
 ## Reporting compatibility problems
 

package/README.md

@@ -2,14 +2,13 @@
 
 > Teach your AI coding assistant to work like a senior QA engineer — and stop it from telling you tests pass when they don't.
 
-[![Version](https://img.shields.io/badge/version-0.9.0-blue.svg)](CHANGELOG.md)
+[![Version](https://img.shields.io/badge/version-0.10.0-blue.svg)](CHANGELOG.md)
 [![Status](https://img.shields.io/badge/status-public%20preview-orange.svg)](docs/release/v0.9-release-checklist.md)
 [![License: MIT](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)
-[![Node](https://img.shields.io/badge/node-%E2%89%A518.18-339933.svg)](#step-1--check-your-prerequisites)
-[![Python](https://img.shields.io/badge/python-%E2%89%A53.8-3776AB.svg)](#step-1--check-your-prerequisites)
+[![Node](https://img.shields.io/badge/node-%E2%89%A518.18-339933.svg)](#step-1--check-your-prerequisite)
 [![Agent Skills](https://img.shields.io/badge/Agent%20Skills-spec--native-6E56CF.svg)](https://agentskills.io)
 [![CI](https://github.com/abisheik88/qa-engineer/actions/workflows/ci.yml/badge.svg)](https://github.com/abisheik88/qa-engineer/actions/workflows/ci.yml)
-[![Tests](https://img.shields.io/badge/tests-235%20passing-success.svg)](#how-this-is-verified)
+[![Tests](https://img.shields.io/badge/tests-165%20passing-success.svg)](#how-this-is-verified)
 
 **This page is everything you need.** Install, first run, every command, and what to do when something breaks — no other document required.
 
@@ -19,7 +18,7 @@
 
 1. [What this is, in plain language](#what-this-is-in-plain-language)
 2. [Is this for me?](#is-this-for-me)
-3. [Step 1 — Check your prerequisites](#step-1--check-your-prerequisites)
+3. [Step 1 — Check your prerequisites](#step-1--check-your-prerequisite)
 4. [Step 2 — Install it](#step-2--install-it)
 5. [Step 3 — Confirm it worked](#step-3--confirm-it-worked)
 6. [Step 4 — Your first real task](#step-4--your-first-real-task)
@@ -67,21 +66,26 @@ You do not need to learn a new language, framework, or config file. You install
 
 **Works best with Playwright** today. Selenium, Cypress, and WebdriverIO are detected and their results are understood, but running and generating tests live is currently Playwright-only. See [framework support](#framework-support).
 
-## Step 1 — Check your prerequisites
+## Step 1 — Check your prerequisite
 
-You need two things. Run these two commands to check — copy and paste them exactly:
+You need one thing. Run this to check — copy and paste it exactly:
 
 ```bash
 node --version
-python3 --version
 ```
 
-You want **Node 18.18 or newer** and **Python 3.8 or newer**. If both print a version number that is high enough, skip ahead to [Step 2](#step-2--install-it).
+You want **Node 18.18 or newer**. If it prints a high enough version, skip ahead to
+[Step 2](#step-2--install-it).
+
+That is the whole list. There is nothing to `pip install`, no second language, and no
+dependencies: the tools that read your test results are dependency-free JavaScript
+that runs on the same Node you just checked.
 
 <details>
 <summary><b>"command not found" — how to install Node.js</b></summary>
 
-Node.js gives you the `npx` command used to install this pack.
+Node.js gives you the `npx` command used to install this pack, and it runs the
+analysis tools afterwards.
 
 - **macOS** — `brew install node` (needs [Homebrew](https://brew.sh)), or download from [nodejs.org](https://nodejs.org).
 - **Windows** — download the LTS installer from [nodejs.org](https://nodejs.org) and run it.
@@ -91,19 +95,6 @@ Then re-run `node --version`. Take the **LTS** version if offered a choice.
 
 </details>
 
-<details>
-<summary><b>"command not found" — how to install Python</b></summary>
-
-Python runs the analysis tools that read your test results. It uses only Python's built-in library — there is nothing extra to install, no `pip install` step.
-
-- **macOS** — `brew install python3`, or download from [python.org](https://python.org).
-- **Windows** — install from [python.org](https://python.org) and **tick "Add Python to PATH"** during setup. Then use `python --version` if `python3` is not found.
-- **Linux** — `sudo apt install python3` (Debian/Ubuntu).
-
-**Can you skip Python?** Yes, but you shouldn't. Without it the skills fall back to the AI reading files by eye, and they will mark their results *degraded* to tell you so. The whole point of this project is that the numbers come from tools, not guesses.
-
-</details>
-
 You also need **an AI coding assistant** open on your project — Claude Code, Cursor, GitHub Copilot, Codex CLI, OpenCode, Gemini CLI, Antigravity, or Kimi.
 
 ## Step 2 — Install it
@@ -165,7 +156,7 @@ Real output from a healthy install:
 ✓ [PASS] skills: 13 skill(s) installed
 ✓ [PASS] contracts: 12 contract schema(s) present
 ✓ [PASS] engine: deterministic engine bundled under .agents/skills/qa-init/scripts/lib
-✓ [PASS] python-imports: Python imports OK (python3 Python 3.12.3)
+✓ [PASS] engine-runs: bundled engine runs cleanly
 ✓ [PASS] node: Node v24.18.0
 ✓ self-test PASSED
 ```
@@ -256,9 +247,27 @@ qa-artifacts/
 
 Most teams **do** commit `.agents/skills/` and `qa-lock.json`, so everyone on the team gets the same skills.
 
-## All twelve commands
+## How you type a command in your assistant
+
+Type these in your AI assistant's **chat**, not the terminal. The keystroke differs
+by assistant, and typing the wrong one looks exactly like a failed install — so
+`qa install` prints the right one for whatever it detected. For reference:
 
-Type these in your AI assistant's chat, not the terminal.
+| Assistant | How to reach a skill |
+| --- | --- |
+| Claude Code | `/qa-explore`, or just describe the task |
+| Cursor | type `/` in Agent chat, then pick the skill from the list |
+| OpenAI Codex CLI | `$qa-explore` — or `/skills` to browse them |
+| OpenCode | `/qa-explore`, or describe the task and the agent loads the skill itself |
+| Antigravity | describe the task; skills activate on their own |
+| GitHub Copilot / Gemini CLI | `/qa-explore` |
+
+Every one of them also activates a skill on its own when your request matches — you
+never *have* to remember a command. "Find bugs on my login page at localhost:3000"
+reaches `/qa-explore` in all of them. Full details and the source for each path:
+[COMPATIBILITY.md](COMPATIBILITY.md).
+
+## All twelve commands
 
 | Command | What it does | Changes your files? |
 | --- | --- | --- |
@@ -329,10 +338,9 @@ npx qa-engineer install --yes --force --project .
 
 ### A skill says the engine is missing, or results are "degraded"
 
-The Python tools aren't reachable. Fix in this order:
+The deterministic engine isn't reachable. Fix in this order:
 
 ```bash
-python3 --version                          # is Python installed at all?
 npx qa-engineer repair --project .         # reinstall the bundled tools
 npx qa-engineer doctor --project .         # should now say "bundled engine runs cleanly"
 ```
@@ -391,7 +399,7 @@ The core idea in one line: **tools produce the facts, the AI explains them.**
               │  raw output, exit code
               ▼
   ┌───────────────────────────────────┐
-  │  Bundled Python tools             │  count the results, classify the failure,
+  │  Bundled Node engine              │  count the results, classify the failure,
   │  (installed inside the skill)     │  redact secrets — no guessing
   └───────────────────────────────────┘
               │  facts
@@ -441,11 +449,11 @@ Everything below is reproducible from a clone with the command beside it.
 
 | Evidence | Command |
 | --- | --- |
-| 152 analysis, framework, and branding tests | `python3 shared/analysis/lib/run_tests.py` |
-| 28 diagnostic engine tests · 5 seam tests | `PYTHONPATH=shared/analysis/lib:shared/diagnostics/lib python3 -m unittest discover -s shared/diagnostics/lib/tests` |
+| 82 engine tests: analysis, diagnostics, frameworks, seams | `node --test packages/engine/test/*.test.mjs` |
+| The recorded corpus, proven against a second implementation | `node --test packages/engine/test/corpus.test.mjs` |
 | 50 installer tests — including security and repeated-use stress | `npm test` |
 | 21 evaluation cases, including deliberately dishonest outputs the scorer must reject | `npm run validate:evals` |
-| 4 real AI-produced results, scored | `python3 tests/evals/run_live.py --captures claude-opus-5` |
+| 4 real AI-produced results, scored | `node tests/evals/run-live.mjs --captures claude-opus-5` |
 | 17 repository checks — including "documentation matches implementation" | `npm run validate:skills` … |
 
 ## Going deeper

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "qa-engineer",
-  "version": "0.9.1",
+  "version": "0.10.0",
   "description": "A vendor-neutral Agent Skills pack that teaches AI coding agents to work like senior QA automation engineers.",
   "license": "MIT",
   "type": "module",
@@ -29,42 +29,34 @@
     "packages/installer/schemas",
     "packages/installer/package.json",
     "packages/installer/README.md",
+    "packages/engine/lib",
+    "packages/engine/bin",
+    "packages/engine/package.json",
     "skills",
-    "shared/analysis/lib/qa_analysis",
-    "shared/analysis/schemas",
-    "shared/diagnostics/lib/qa_diagnostics",
-    "shared/diagnostics/schemas",
-    "shared/tooling/qa_tool.py",
     "shared/frameworks/registry.json",
     "shared/frameworks/registry.schema.json",
     "shared/frameworks/registry.mjs",
-    "shared/frameworks/playwright/lib/playwright_analysis.py",
-    "shared/frameworks/selenium/lib/selenium_analysis.py",
-    "shared/frameworks/cypress/lib/cypress_analysis.py",
-    "shared/frameworks/webdriverio/lib/webdriverio_analysis.py",
     "COMPATIBILITY.md",
-    "!**/__pycache__",
-    "!**/*.pyc",
-    "!**/tests"
+    "!**/tests",
+    "shared/tooling/qa-tool.mjs",
+    "!packages/engine/test"
   ],
   "scripts": {
+    "eval:live": "node tests/evals/run-live.mjs --baseline tests/evals/baselines/reference.json",
+    "prepublishOnly": "npm run validate:release && npm run validate:doc-claims && npm run validate:docs-commands && npm test && npm run validate:evals",
     "qa": "node packages/installer/bin/qa.mjs",
-    "eval:live": "python3 tests/evals/run_live.py --baseline tests/evals/baselines/reference.json",
-    "prepublishOnly": "npm run validate:release && npm run validate:doc-claims && npm run validate:docs-commands && npm test && npm run validate:python && npm run validate:evals",
     "release:checksums": "node scripts/release/release-notes.mjs --checksums",
     "release:notes": "node scripts/release/release-notes.mjs",
-    "test": "node --test packages/installer/test/*.test.mjs",
+    "test": "node --test packages/installer/test/*.test.mjs packages/engine/test/*.test.mjs",
     "validate:architecture": "node scripts/check-architecture-fitness.mjs",
     "validate:branding": "node scripts/check-branding.mjs",
-    "validate:bundle": "python3 scripts/bundle_python.py --check",
-    "validate:context": "cd shared/analysis/lib && python3 -m qa_analysis.cli context --path tests/fixtures/valid-context.md",
     "validate:doc-claims": "node scripts/check-doc-claims.mjs",
     "validate:docs-commands": "node scripts/check-docs-commands.mjs",
-    "validate:evals": "python3 tests/evals/run_evals.py",
+    "validate:engine": "node --test packages/engine/test/*.test.mjs",
+    "validate:evals": "node tests/evals/run-evals.mjs",
     "validate:keywords": "node scripts/check-keywords.mjs",
     "validate:knowledge": "node scripts/check-knowledge.mjs",
     "validate:matrix": "node scripts/check-capability-matrix.mjs",
-    "validate:python": "python3 shared/analysis/lib/run_tests.py",
     "validate:registry": "node scripts/check-framework-registry.mjs",
     "validate:release": "node scripts/release/validate-release.mjs",
     "validate:release-notes": "node scripts/release/release-notes.mjs --check",