Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Insajin/autopus-adk --skill testing-strategygit clone --depth 1 https://github.com/Insajin/autopus-adkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/insajin/autopus-adk/testing-strategy)<a href="https://agentmods.dev/skills/insajin/autopus-adk/testing-strategy"><img src="https://agentmods.dev/badge/skills/insajin/autopus-adk/testing-strategy.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00020 | $0.01480 |
| Opus 5 | $0.00010 | $0.00740 |
| Sonnet 5 | $0.00004 | $0.00296 |
| Haiku 4.5 | $0.00002 | $0.00148 |
Grade A, and why
testing-strategy scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Testing Strategy Skill
단위 테스트를 넘어 통합/E2E/계약 테스트를 설계하는 스킬입니다.
테스트 피라미드
/ E2E \ ← 적게, 핵심 플로우만
/ 통합 \ ← 중간, 컴포넌트 간 연동
/ 단위 \ ← 많이, 빠르고 격리
| 계층 | 비중 | 속도 | 범위 |
|---|---|---|---|
| 단위 | 70% | 빠름 (ms) | 함수/메서드 |
| 통합 | 20% | 중간 (s) | 컴포넌트 간 |
| E2E | 10% | 느림 (min) | 전체 시스템 |
통합 테스트
데이터베이스 통합
func TestUserRepo_Integration(t *testing.T) {
if testing.Short() {
t.Skip("통합 테스트 건너뜀")
}
db := setupTestDB(t)
defer db.Close()
repo := NewUserRepo(db)
user, err := repo.Create(ctx, &User{Name: "test"})
require.NoError(t, err)
found, err := repo.FindByID(ctx, user.ID)
require.NoError(t, err)
assert.Equal(t, "test", found.Name)
}
HTTP API 통합
func TestAPI_CreateUser(t *testing.T) {
srv := setupTestServer(t)
defer srv.Close()
resp, err := http.Post(srv.URL+"/api/v1/users",
"application/json",
strings.NewReader(`{"name":"test"}`))
require.NoError(t, err)
assert.Equal(t, http.StatusCreated, resp.StatusCode)
}
E2E 테스트
핵심 플로우만 테스트
사용자 등록 → 로그인 → 리소스 생성 → 조회 → 삭제
QAMESH Source Guidance
- QAMESH is the default project QA orchestration layer. Playwright is not a competing mode; when detected, it is registered as a browser/gui Journey runner adapter.
- Ask users to choose the project under test, execution authority, environment/origin, credentials boundary, mobile/cloud device boundary, or explicit canary command. Do not ask them to choose between QAMESH and Playwright.
- Use
auto qa init --format jsonas the simple default release-QA setup command for arbitrary projects. It creates project-local starter Journey Packs plus a generic GitHub Actions QAMESH release gate. - Use
auto qa init --local-only --format jsonwhen only Journey Pack starters are needed. Review generated commands, env, installer version, and required-gate policy before trusting the workflow. - Use
auto qa plan --format jsonbefore project-level QA execution to inspect Journey Packs, detected adapters, selected lanes, setup gaps, and output paths without running commands. - Use
auto canaryonly for post-deploy smoke/status verification. QAMESH owns deterministic user journey evidence, redacted artifacts, run indexes, and repair feedback;auto qa releasetreatscanary-explicitas a bridge lane for an explicit post-deploy smoke Journey Pack. - Use
auto qa run --format jsonwhen deterministic project QA should execute and produce QAMESH run/evidence output. - Use
auto qa explore --dry-run --format jsonbefore GUI exploration; execute it only for explicit local/staging Journey Packs with allowed origins, forbidden actions, deterministic oracles, and redacted artifact retention. - Use
auto qa scenario compile --format jsonto turn project-declared user scenarios in.autopus/qa/scenarios/*.yamlinto Playwright specs; the step vocabulary is read-only by construction, so a compiled scenario cannot trip the GUI forbidden-action guard. - Use
auto qa release --dry-run --format jsonto inspect the fixed release lane set, setup gaps, blocker matrix, redacted command previews, and sibling SPEC readiness before launch gates; useauto qa release --roadmap --format jsonfor the canonical roadmap surface. - Use
auto qa evidencewhen an external producer already wrote a QAMESH manifest and the task is validation, redaction, and publication. - Use
auto qa feedbackto turn existing failed QAMESH evidence into provider-specific repair prompt bundles. - ADK is a harness: each concrete Journey Pack is a project-local Journey Pack under
.autopus/qa/journeys/**, while ADK owns adapters, execution, redaction, and feedback plumbing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 142 lines · 20 tokens per session scan A ef0e448a28c1
testing-strategy is a skill published in the GitHub repository Insajin/autopus-adk (110 stars, last pushed today), licensed MIT. It adds 20 tokens to every session and 1,480 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
ui-testing
Create automated UI tests using Playwright for Streamlit and web applications. Use when writing end-to-end tests, automating UI testing, or testing new features.
playwright
Complete browser automation with Playwright. Auto-detects dev servers, writes clean test scripts to /tmp. Test pages, fill forms, take screenshots, check responsive design, validate UX, test login flows, check links, automate any browser task. Use when user wants to test websites, automate browser interactions…
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
test-reporting
Run the Level 2 dummy agent integration test suite and produce a detailed HTML report with per-test input → outcome analysis.
android-emulator
Verify and debug native, React Native, Expo, or Flutter apps on an Android Emulator with agent-device. Use when an agent needs to launch an app, inspect its live UI, tap, type, scroll, validate a code change, collect failure evidence, or reproduce a workflow on an Android virtual device.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.