Adoption

Agent Skills are supported by leading AI development tools.

VS Code Gemini CLI GitHub Goose Amp Cursor Claude Code Letta OpenCode Claude OpenAI Codex Factory VS Code Gemini CLI GitHub Goose Amp Cursor Claude Code Letta OpenCode Claude OpenAI Codex Factory

astra-technology-company-limited/uat-parallel

Name: uat-parallel
Author: astra-technology-company-limited

skills/uat-parallel/SKILL.md

npx skillsauth add astra-technology-company-limited/astra-methodology uat-parallel

Clean

TrivyContainer and dependency vulnerability scanner

Clean

SemgrepStatic code analysis for vulnerabilities

Clean

mcp-scan (Snyk)Model Context Protocol security validation

Skipped

Snyk (dep)Open source security scanning

Skipped

Socket.devSupply chain security analysis

Skipped

VirusTotalMulti-engine malware detection

Skipped

CrowdStrikeAdvanced threat intelligence

Skipped

OSV-ScannerOpen Source Vulnerability database check

Skipped

OWASP Dep-Check

ASTRA /uat-parallel — Playwright-powered Parallel UAT

Runs UAT cases concurrently using Playwright Test workers. Each worker holds an isolated browser context, so authentication, cookies, and localStorage do not leak between cases. Cases that hang are killed by per-step / per-case timeouts and never block other workers.

This skill is the parallel sibling of /user-test. Use /user-test for interactive ad-hoc testing or when you do not have Playwright installed; use /uat-parallel for batch regression runs.

| Aspect | /user-test | /uat-parallel | |---|---|---| | Driver | Chrome MCP (in-Claude tool calls) | Playwright CLI (subprocess) | | Concurrency | 1 page, sequential | N workers, parallel | | Browser state | Shared across cases | Isolated per worker | | Per-case timeout | None (Claude must self-abort) | Hard timeout enforced by Playwright | | Interactive mode | Yes (URL + Vietnamese flow) | No (auto-batch only) | | Trace / replay | Static screenshots only | Playwright trace viewer (.zip) | | Bootstrap cost | Zero | One-time npm i -D @playwright/test | | Case format | docs/tests/uat-cases/*.md | Same files, no migration |

Detailed references (load on demand):

references/parallel-guide.md — worker count tuning, isolation guarantees, debugging flaky cases.
../user-test/references/assertion-guide.md — assertion grammar and severity rules (reused as-is).
../user-test/references/i18n-strings.md — vi/en/ko translation table (shared SSoT; the merge script reads UAT_LANG and renders accordingly).

1. Bootstrap (run once per project)

Check whether Playwright is available:

npx --no-install playwright --version 2>/dev/null

If the command exits non-zero or package.json does not contain @playwright/test:

Ask the user via AskUserQuestion:

"Skill này cần Playwright để chạy song song. Cài @playwright/test + Chromium ngay không? (~150 MB)"
- Có (Recommended) → run installation
- Không → abort with message: Đã huỷ. Chạy thủ công: npm i -D @playwright/test && npx playwright install chromium

On confirmation:

# In project root
if [ ! -f package.json ]; then npm init -y >/dev/null; fi
npm i -D @playwright/test
npx playwright install chromium

Verify: npx playwright --version must succeed.

Once installed, skip bootstrap on subsequent runs.

2. Argument parsing

| Flag | Default | Meaning | |---|---|---| | --workers N | min(4, max(1, cpu/2)) | Number of parallel workers | | --from {glob} | docs/tests/uat-cases/*.md | Restrict case files | | --priority X | none | Filter by frontmatter priority | | --feature Y | none | Filter by frontmatter feature | | --timeout 30s | 30s per step, 300s per case | Hard timeouts | | --headed | false (headless) | Show browser windows | | --browser X | chromium | chromium | firefox | webkit | | --lang X | resolved at Step 0 | Report-output language (vi | en | ko) |

3. Pipeline

Step 0 — Language Selection (report output language)

Determine LANG_CODE ∈ {vi, en, ko} — used for index.html (<html lang> + visible labels), issues.md headings, and console log messages. UAT case file contents are not translated.

Resolution order:

--lang flag in $ARGUMENTS → normalize case-insensitive (vi|vie|vietnamese → vi; en|eng|english → en; ko|kor|korean → ko). If recognized, skip to Step 1.
Persisted CLAUDE.md ## Language → if it resolves to ko, vi, or en, use it silently.
Otherwise → ask via AskUserQuestion with the trilingual prompt below; default selection Vietnamese.

Chọn ngôn ngữ cho báo cáo UAT.
Select the language for the UAT report.
UAT 보고서 언어를 선택하세요.

Options (single-select, header Lang):

Tiếng Việt — Vietnamese (Recommended)
English — English
한국어 — Korean

Map: Tiếng Việt → vi, English → en, 한국어 → ko.

The merge script (uat-parallel-report.sh) consumes LANG_CODE via the UAT_LANG environment variable in Step 5.

Step 1 — Validate and select cases

Glob case files using --from (default docs/tests/uat-cases/*.md).
Read YAML frontmatter from each; apply --priority and --feature filters.
If zero cases match → abort with: Không tìm thấy UAT case phù hợp.
Announce: Tìm thấy {N} test case, sẽ chạy song song với {W} workers.

Step 2 — Initialize session folder

SESSION_ID  = {YYYY-MM-DD-HHmm}
SESSION_DIR = docs/tests/uat-reports/{SESSION_ID}/

Create:

{SESSION_DIR}/screenshots/
{SESSION_DIR}/traces/
{SESSION_DIR}/raw/ (Playwright JSON output lands here)

Step 3 — Install runner files (idempotent)

The runner files live at .astra/uat/ in the project. On every invocation, copy from the plugin (overwrite — these are managed files):

| Source (plugin) | Destination (project) | |---|---| | $CLAUDE_PLUGIN_ROOT/skills/uat-parallel/assets/uat-runner.spec.ts | .astra/uat/uat-runner.spec.ts | | $CLAUDE_PLUGIN_ROOT/skills/uat-parallel/assets/playwright.config.ts | .astra/uat/playwright.config.ts |

Add .astra/uat/ to .gitignore if missing.

Step 4 — Run Playwright

Set environment variables that the runner spec reads, then invoke:

cd <project-root>
UAT_SESSION_DIR="{absolute SESSION_DIR}" \
UAT_CASES_GLOB="{--from glob or default}" \
UAT_FILTER_PRIORITY="{--priority value or empty}" \
UAT_FILTER_FEATURE="{--feature value or empty}" \
UAT_STEP_TIMEOUT_MS="{step ms}" \
UAT_CASE_TIMEOUT_MS="{case ms}" \
npx playwright test \
  --config .astra/uat/playwright.config.ts \
  --workers {N} \
  --browser {browser} \
  {--headed if requested} \
  --reporter json,html \
  --output {SESSION_DIR}/raw

The runner spec writes per-case results into {SESSION_DIR}/raw/results/{case-id}.json plus screenshots into {SESSION_DIR}/screenshots/{case-id}/step-NN.png. Failed cases also produce traces at {SESSION_DIR}/traces/{case-id}.zip.

Do not abort on non-zero exit — Playwright returns non-zero whenever any case FAILs, which is normal.

Step 5 — Merge results

Invoke the merge helper. Pass UAT_LANG (the LANG_CODE from Step 0) and UAT_WORKERS (worker count) as environment variables so the script can localize HTML + issues.md:

UAT_LANG="{LANG_CODE}" \
UAT_WORKERS="{N}" \
bash $CLAUDE_PLUGIN_ROOT/skills/uat-parallel/scripts/uat-parallel-report.sh \
  "{SESSION_DIR}" \
  "$CLAUDE_PLUGIN_ROOT/skills/user-test/assets/report-template.html"

The script:

Reads every {SESSION_DIR}/raw/results/*.json.
Aggregates pass/fail counts, durations, severities.
Renders {SESSION_DIR}/index.html from the user-test template (reused) — substitutes {{SESSION_ID}}, {{MODE}} = "auto-parallel (W workers)", summary stats, {{TEST_CASES_HTML}}, {{ISSUES_HTML}}.

Writes {SESSION_DIR}/issues.md if any FAIL exists, using the same schema as /user-test:

## Issue #N — {SEVERITY}
**Test Case**: {UAT-ID} - {name}
**Bước**: {step_num}/{total} — {step_name}
### Expected ...
### Actual ...
### Lý do gán {SEVERITY}
### Screenshot
### Trace (Playwright)
`npx playwright show-trace ./traces/{case-id}.zip`

Writes consolidated {SESSION_DIR}/session.json with summary, cases, issues.
Removes {SESSION_DIR}/raw/ after a successful merge.

Step 6 — Summarize

Print using the L_DONE_PARALLEL string for the resolved LANG_CODE. Example for vi:

▶ Hoàn thành (parallel, {W} workers).
   📊 {PASS} PASS / {FAIL} FAIL  ·  {DURATION}
   📄 {SESSION_DIR}/index.html
   🐛 issues.md có {M} issue ({severity breakdown})
   🎞️  Trace replay: npx playwright show-trace {SESSION_DIR}/traces/{first-failed}.zip

For en use ▶ Done (parallel, {W} workers) + issues.md has {M} issues; for ko use ▶ 완료 (병렬, {W} workers) + issues.md에 {M}건의 이슈. Skip the issues.md and trace lines when no FAIL.

4. UAT case file format

100% identical to /user-test — no migration required. See skills/user-test/SKILL.md §3. Recap:

---
id: UAT-001
name: Đăng ký tài khoản
priority: critical
feature: auth
base_url: https://staging.fect.app
---

## Bước 1: Mở trang chủ
**Action**: navigate `{base_url}`
**Expected**:
- url: equals `{base_url}/`
- dom: `button:has-text("Đăng ký")` exists

Action verbs (navigate, click, fill, wait, press, hover) and assertion kinds (url, network, dom, console) are interpreted by .astra/uat/uat-runner.spec.ts. The grammar reference is shared with /user-test at skills/user-test/references/assertion-guide.md.

5. Output structure

docs/tests/uat-reports/2026-05-29-1830/
├── index.html              # Same look as /user-test report
├── issues.md               # Only if FAIL > 0
├── session.json
├── screenshots/
│   └── UAT-001/
│       ├── step-01.png
│       └── step-02.png
└── traces/                 # Only for failed cases
    └── UAT-001.zip

6. Examples

# Run all cases with default workers
/uat-parallel

# Run with 6 workers, only critical priority
/uat-parallel --workers 6 --priority critical

# Single case, headed mode for visual debugging
/uat-parallel --from docs/tests/uat-cases/dang-ky.md --headed --workers 1

# Cross-browser regression
/uat-parallel --browser firefox --workers 3

# Generate report in English / Korean
/uat-parallel --lang en
/uat-parallel --workers 6 --priority critical --lang ko

7. Standing instructions

User-facing output language: resolved at Step 0 (--lang flag → CLAUDE.md ## Language → AskUserQuestion → default vi). Passed to the merge script via UAT_LANG. The script embeds an inline mirror of the skills/user-test/references/i18n-strings.md dictionary (the SSoT file is documentation; the runtime values live inside the Node heredoc) and substitutes every visible string in index.html + issues.md accordingly. When adding or changing strings, update both the SSoT and the script. /uat-parallel does not render M_DEV_HINT (the Playwright runner has no LLM to author per-failure hints) and uses M_ISSUES_REPORT_TITLE_PARALLEL instead of the base title key. File slugs always use ASCII.
Reuse /user-test assets: do NOT duplicate the HTML template or assertion grammar — load from skills/user-test/ paths. Future updates to /user-test's template propagate automatically.
Hard assertions only: same rule as /user-test. URL / Network / DOM / Console only.
Severity rules are shared: identical to references/assertion-guide.md §3. The runner emits the raw failure; the merge script applies the severity rules.
Per-case isolation is the value prop: never share BrowserContext across cases. The runner enforces this via Playwright's per-test context fixture.
Timeout is non-negotiable: a stuck case must die, not block. Default 30s/step, 300s/case. Caller can override via --timeout.
No GitHub Issue integration: write only to local issues.md, same as /user-test v2+.
Idempotent runner install: copy .astra/uat/ files from the plugin on every run — they are managed artifacts, not user-editable.

8. Anti-scope (do NOT use for)

Interactive ad-hoc UAT (URL + Vietnamese flow on the fly) → use /user-test.
Authoring new UAT cases from a blueprint → use /test-scenario first.
Developer technical scenarios → use /test-run.
Pure API testing (no UI) → use curl / integration tests.
Unit tests → use the project's test framework (Jest/JUnit/pytest).
Visual regression (pixel diff) → Playwright toHaveScreenshot() is supported by Playwright itself, but this skill enforces hard assertions only.

9. Failure modes & remedies

| Symptom | Likely cause | Remedy | |---|---|---| | Bootstrap fails: EACCES on npm i | Permission on node_modules/ | Run sudo chown -R $USER node_modules | | npx playwright install hangs | Network blocked | Set HTTPS_PROXY or pre-download browsers | | All cases fail with "timeout waiting for selector" | base_url unreachable | Verify dev server, prefer staging URL in frontmatter | | Worker count higher than CPU → slower run | Context switching overhead | Lower --workers to cpu/2 | | Auth state leaks across cases | Manual context reuse in spec | Should not happen — re-install .astra/uat/ files from plugin | | Trace files missing | --trace on-first-retry skipped | Re-run with --workers 1 to force trace on first attempt |

astra-technology-company-limited/uat-parallel

skills/uat-parallel/SKILL.md

Runs UAT (User Acceptance Testing) cases in TRUE PARALLEL using Playwright Test runner with isolated browser contexts per worker (separate cookies, localStorage, sessionStorage). Solves the two main limits of /user-test: (1) sequential single-page execution that does not scale beyond a few cases, and (2) one stuck case blocking the rest of the run. Reuses 100% of the /user-test UAT case Markdown+YAML format under docs/tests/uat-cases/, runs them via `npx playwright test --workers=N`, and emits the same report layout (index.html + issues.md + session.json + screenshots/) under docs/tests/uat-reports/. Use when the user asks to "run UAT in parallel", "speed up UAT", "test multi-user", "song song", "uat parallel", or runs /uat-parallel. Distinct from /user-test (sequential Chrome MCP, supports interactive mode), /test-run (developer integration tests), /test-scenario (scenario authoring).

tools

Updated Jun 4, 2026

$ install --global

skillsauth

npx skillsauth add astra-technology-company-limited/astra-methodology uat-parallel

Install this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.

Security Scan Results

3 of 9 scanners reported clean

Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.

Scanners Passed

Scanners in report

Clean

TrivyContainer and dependency vulnerability scanner

95%

Clean

SemgrepStatic code analysis for vulnerabilities

95%

Clean

mcp-scan (Snyk)Model Context Protocol security validation

95%

Skipped

Snyk (dep)Open source security scanning

50%

Skipped

Socket.devSupply chain security analysis

50%

Skipped

VirusTotalMulti-engine malware detection

50%

Skipped

CrowdStrikeAdvanced threat intelligence

50%

Skipped

OSV-ScannerOpen Source Vulnerability database check

50%

Skipped

OWASP Dep-Check

50%

Last scanned: Jun 4, 2026, 5:01 AM271.2s5 files scanned

SKILL.md

name:: uat-parallel
description:: >
argument-hint:: [--workers N] [--from glob] [--priority critical|high|medium|low] [--feature name] [--timeout 30s] [--headed] [--browser chromium|firefox|webkit] [--lang vi|en|ko]
allowed-tools:: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion

ASTRA /uat-parallel — Playwright-powered Parallel UAT

This skill is the parallel sibling of /user-test. Use /user-test for interactive ad-hoc testing or when you do not have Playwright installed; use /uat-parallel for batch regression runs.

Detailed references (load on demand):

references/parallel-guide.md — worker count tuning, isolation guarantees, debugging flaky cases.
../user-test/references/assertion-guide.md — assertion grammar and severity rules (reused as-is).
../user-test/references/i18n-strings.md — vi/en/ko translation table (shared SSoT; the merge script reads UAT_LANG and renders accordingly).

1. Bootstrap (run once per project)

Check whether Playwright is available:

npx --no-install playwright --version 2>/dev/null

If the command exits non-zero or package.json does not contain @playwright/test:

Ask the user via AskUserQuestion:

"Skill này cần Playwright để chạy song song. Cài @playwright/test + Chromium ngay không? (~150 MB)"
- Có (Recommended) → run installation
- Không → abort with message: Đã huỷ. Chạy thủ công: npm i -D @playwright/test && npx playwright install chromium

On confirmation:

# In project root
if [ ! -f package.json ]; then npm init -y >/dev/null; fi
npm i -D @playwright/test
npx playwright install chromium

Verify: npx playwright --version must succeed.

Once installed, skip bootstrap on subsequent runs.

2. Argument parsing

3. Pipeline

Step 0 — Language Selection (report output language)

Determine LANG_CODE ∈ {vi, en, ko} — used for index.html (<html lang> + visible labels), issues.md headings, and console log messages. UAT case file contents are not translated.

Resolution order:

--lang flag in $ARGUMENTS → normalize case-insensitive (vi|vie|vietnamese → vi; en|eng|english → en; ko|kor|korean → ko). If recognized, skip to Step 1.
Persisted CLAUDE.md ## Language → if it resolves to ko, vi, or en, use it silently.
Otherwise → ask via AskUserQuestion with the trilingual prompt below; default selection Vietnamese.

Chọn ngôn ngữ cho báo cáo UAT.
Select the language for the UAT report.
UAT 보고서 언어를 선택하세요.

Options (single-select, header Lang):

Tiếng Việt — Vietnamese (Recommended)
English — English
한국어 — Korean

Map: Tiếng Việt → vi, English → en, 한국어 → ko.

The merge script (uat-parallel-report.sh) consumes LANG_CODE via the UAT_LANG environment variable in Step 5.

Step 1 — Validate and select cases

Glob case files using --from (default docs/tests/uat-cases/*.md).
Read YAML frontmatter from each; apply --priority and --feature filters.
If zero cases match → abort with: Không tìm thấy UAT case phù hợp.
Announce: Tìm thấy {N} test case, sẽ chạy song song với {W} workers.

Step 2 — Initialize session folder

SESSION_ID  = {YYYY-MM-DD-HHmm}
SESSION_DIR = docs/tests/uat-reports/{SESSION_ID}/

Create:

{SESSION_DIR}/screenshots/
{SESSION_DIR}/traces/
{SESSION_DIR}/raw/ (Playwright JSON output lands here)

Step 3 — Install runner files (idempotent)

The runner files live at .astra/uat/ in the project. On every invocation, copy from the plugin (overwrite — these are managed files):

Add .astra/uat/ to .gitignore if missing.

Step 4 — Run Playwright

Set environment variables that the runner spec reads, then invoke:

cd <project-root>
UAT_SESSION_DIR="{absolute SESSION_DIR}" \
UAT_CASES_GLOB="{--from glob or default}" \
UAT_FILTER_PRIORITY="{--priority value or empty}" \
UAT_FILTER_FEATURE="{--feature value or empty}" \
UAT_STEP_TIMEOUT_MS="{step ms}" \
UAT_CASE_TIMEOUT_MS="{case ms}" \
npx playwright test \
  --config .astra/uat/playwright.config.ts \
  --workers {N} \
  --browser {browser} \
  {--headed if requested} \
  --reporter json,html \
  --output {SESSION_DIR}/raw

Do not abort on non-zero exit — Playwright returns non-zero whenever any case FAILs, which is normal.

Step 5 — Merge results

Invoke the merge helper. Pass UAT_LANG (the LANG_CODE from Step 0) and UAT_WORKERS (worker count) as environment variables so the script can localize HTML + issues.md:

UAT_LANG="{LANG_CODE}" \
UAT_WORKERS="{N}" \
bash $CLAUDE_PLUGIN_ROOT/skills/uat-parallel/scripts/uat-parallel-report.sh \
  "{SESSION_DIR}" \
  "$CLAUDE_PLUGIN_ROOT/skills/user-test/assets/report-template.html"

The script:

Reads every {SESSION_DIR}/raw/results/*.json.
Aggregates pass/fail counts, durations, severities.
Renders {SESSION_DIR}/index.html from the user-test template (reused) — substitutes {{SESSION_ID}}, {{MODE}} = "auto-parallel (W workers)", summary stats, {{TEST_CASES_HTML}}, {{ISSUES_HTML}}.

Writes {SESSION_DIR}/issues.md if any FAIL exists, using the same schema as /user-test:

## Issue #N — {SEVERITY}
**Test Case**: {UAT-ID} - {name}
**Bước**: {step_num}/{total} — {step_name}
### Expected ...
### Actual ...
### Lý do gán {SEVERITY}
### Screenshot
### Trace (Playwright)
`npx playwright show-trace ./traces/{case-id}.zip`

Writes consolidated {SESSION_DIR}/session.json with summary, cases, issues.
Removes {SESSION_DIR}/raw/ after a successful merge.

Step 6 — Summarize

Print using the L_DONE_PARALLEL string for the resolved LANG_CODE. Example for vi:

▶ Hoàn thành (parallel, {W} workers).
   📊 {PASS} PASS / {FAIL} FAIL  ·  {DURATION}
   📄 {SESSION_DIR}/index.html
   🐛 issues.md có {M} issue ({severity breakdown})
   🎞️  Trace replay: npx playwright show-trace {SESSION_DIR}/traces/{first-failed}.zip

4. UAT case file format

100% identical to /user-test — no migration required. See skills/user-test/SKILL.md §3. Recap:

---
id: UAT-001
name: Đăng ký tài khoản
priority: critical
feature: auth
base_url: https://staging.fect.app
---

## Bước 1: Mở trang chủ
**Action**: navigate `{base_url}`
**Expected**:
- url: equals `{base_url}/`
- dom: `button:has-text("Đăng ký")` exists

5. Output structure

docs/tests/uat-reports/2026-05-29-1830/
├── index.html              # Same look as /user-test report
├── issues.md               # Only if FAIL > 0
├── session.json
├── screenshots/
│   └── UAT-001/
│       ├── step-01.png
│       └── step-02.png
└── traces/                 # Only for failed cases
    └── UAT-001.zip

6. Examples

# Run all cases with default workers
/uat-parallel

# Run with 6 workers, only critical priority
/uat-parallel --workers 6 --priority critical

# Single case, headed mode for visual debugging
/uat-parallel --from docs/tests/uat-cases/dang-ky.md --headed --workers 1

# Cross-browser regression
/uat-parallel --browser firefox --workers 3

# Generate report in English / Korean
/uat-parallel --lang en
/uat-parallel --workers 6 --priority critical --lang ko

7. Standing instructions

User-facing output language: resolved at Step 0 (--lang flag → CLAUDE.md ## Language → AskUserQuestion → default vi). Passed to the merge script via UAT_LANG. The script embeds an inline mirror of the skills/user-test/references/i18n-strings.md dictionary (the SSoT file is documentation; the runtime values live inside the Node heredoc) and substitutes every visible string in index.html + issues.md accordingly. When adding or changing strings, update both the SSoT and the script. /uat-parallel does not render M_DEV_HINT (the Playwright runner has no LLM to author per-failure hints) and uses M_ISSUES_REPORT_TITLE_PARALLEL instead of the base title key. File slugs always use ASCII.
Reuse /user-test assets: do NOT duplicate the HTML template or assertion grammar — load from skills/user-test/ paths. Future updates to /user-test's template propagate automatically.
Hard assertions only: same rule as /user-test. URL / Network / DOM / Console only.
Severity rules are shared: identical to references/assertion-guide.md §3. The runner emits the raw failure; the merge script applies the severity rules.
Per-case isolation is the value prop: never share BrowserContext across cases. The runner enforces this via Playwright's per-test context fixture.
Timeout is non-negotiable: a stuck case must die, not block. Default 30s/step, 300s/case. Caller can override via --timeout.
No GitHub Issue integration: write only to local issues.md, same as /user-test v2+.
Idempotent runner install: copy .astra/uat/ files from the plugin on every run — they are managed artifacts, not user-editable.

8. Anti-scope (do NOT use for)

Interactive ad-hoc UAT (URL + Vietnamese flow on the fly) → use /user-test.
Authoring new UAT cases from a blueprint → use /test-scenario first.
Developer technical scenarios → use /test-run.
Pure API testing (no UI) → use curl / integration tests.
Unit tests → use the project's test framework (Jest/JUnit/pytest).
Visual regression (pixel diff) → Playwright toHaveScreenshot() is supported by Playwright itself, but this skill enforces hard assertions only.

9. Failure modes & remedies

Related Skills

astra-technology-company-limited/user-test

tools

VerifiedTrustedCommunity

Performs end-user UAT (User Acceptance Testing) by driving a real browser through Chrome MCP, self-verifying each step with hard assertions (DOM / Network / URL / Console), auto-assigning severity on failure, and emitting an HTML report plus issues.md into a timestamped session folder. Supports two modes: interactive (URL + Vietnamese natural-language flow description) and --auto (batch-run pre-authored test cases under docs/tests/uat-cases/). Use when the user asks for "UAT", "user acceptance test", "kiểm thử người dùng", "regression test", or runs /user-test, /uat. Distinct from /test-run (developer-authored technical integration testing) and /test-scenario (scenario authoring from blueprints).

SKILL.mdUpdated May 22, 2026

astra-technology-company-limited/user-test

astra-technology-company-limited/tool-author

tools

VerifiedTrustedCommunity

Authors and validates LLM tool descriptions and input schemas (Anthropic Tool Use, MCP servers, LangChain @tool, Pydantic, Zod). Use when the user mentions "tool description", "function calling", "MCP tool", "Pydantic schema", "Zod schema", "@tool decorator", "input_schema", "tool spec", "툴 정의", "함수 호출 스키마", or when editing files that define LLM tool surfaces. Enforces the six required attributes (one-line summary, anti-pattern, synonyms, parameter examples, enum constraints, return shape) and blocks the seven known failure modes — wrong-tool selection, skipped tool, malformed arguments, retry loops, user-intent bypass, wrong side-effect, and un-auditable traces. For authoring ASTRA SKILL.md files use /skill-author instead — this skill is for *runtime* LLM tool surfaces, not for skill files themselves.

SKILL.mdUpdated May 22, 2026

astra-technology-company-limited/tool-author

astra-technology-company-limited/skill-author

development

VerifiedTrustedCommunity

Creates new SKILL.md files or refactors existing skills to comply with the ASTRA skill best practices guide (docs/development/skill-best-practices.md). Use when user mentions "new skill", "create skill", "SKILL.md", "skill authoring", "스킬 작성", "스킬 만들기", or when editing any file matching skills/**/SKILL.md.

SKILL.mdUpdated May 21, 2026

astra-technology-company-limited/skill-author

astra-technology-company-limited/design-redesign

tools

VerifiedTrustedCommunity

Audits existing UI components/pages/CSS against docs/design-system/DESIGN.md and applies fixes to restore design consistency. Uses design-token-validator to detect hardcoded color/font/spacing violations, and invokes designer-persona to report a senior-perspective score (0-10) and anti-AI aesthetic violations (generic shadcn look, purple gradient cliché). Fixes are applied automatically (--apply) or proposed as a PR (--pr); after changes, design-token-validator is re-run and must PASS before completion. Input: target directory (e.g., src/components/Button), single file, or git diff. Output: audit report (docs/design-system/audit-{date}.md) + fix proposals + application results.

SKILL.mdUpdated May 21, 2026

astra-technology-company-limited/design-redesign

Download

For Claude Desktop. Download once, then upload the file in the app — no terminal needed.

Need help? View full Cowork setup guide →

Install manually

Choose your platform

# Clone the repo
git clone https://github.com/astra-technology-company-limited/astra-methodology.git

# Copy into Claude Code skills folder (global)
cp -r astra-methodology/skills/uat-parallel ~/.claude/skills/

Claude Code Skills — official skills path docs.

Repository

astra-technology-company-limited/astra-methodology

Compatible with

Claude Code

OpenAI Codex CLI

ChatGPT