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 agentmods add skills/g-research/bobbit/qa-testnpx skills add G-Research/bobbit --skill qa-testgit clone --depth 1 https://github.com/G-Research/bobbitWrote 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/g-research/bobbit/qa-test)<a href="https://agentmods.dev/skills/g-research/bobbit/qa-test"><img src="https://agentmods.dev/badge/skills/g-research/bobbit/qa-test.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00016 | $0.03425 |
| Opus 5 | $0.00008 | $0.01713 |
| Sonnet 5 | $0.00003 | $0.00685 |
| Haiku 4.5 | $0.00002 | $0.00343 |
Grade C, and why
qa-test scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
2. Remove the temp directory: `rm -rf "$WORK_DIR"` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
if curl -sf "<health_check_url>" > /dev/null 2>&1; then How it starts
The opening of the file, as written. The whole thing — 288 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QA Testing Protocol
You are running QA testing for a goal. This protocol stands up an isolated copy of the application, drives a real browser through user scenarios, captures screenshot evidence, and produces an HTML validation report.
Prerequisites
- You need browser tools available. Use the native browser tools — NOT the
mcp__playwright__*tools. The MCP Playwright browser is a single shared instance across all sessions — other agents and the dev server will hijack your page. The native browser tools give you an isolated browser instance per session. - Available native browser tools:
browser_navigate,browser_screenshot,browser_click,browser_type,browser_eval,browser_wait,browser_snapshot,browser_console_messages,browser_press_key,browser_hover,browser_select_option,browser_resize browser_snapshotis the best way to understand page structure — it returns an ARIA accessibility tree with element roles, names, and refs. Use it instead of screenshots when you need to find interactive elements or verify page content.browser_console_messagescaptures JS console output. Call withlevel="error"after each navigation to catch silent errors.- At least one component in
.bobbit/config/project.yamlmust carryconfig.qa_start_command
Step 1: Read Configuration
Read the project config to discover the component(s) with QA testbed configuration:
cat .bobbit/config/project.yaml
Each component in components[] may carry an opaque config: map. The component you want is the one whose config.qa_start_command is set. Pick it as follows:
- Look for a
[QA-TEST CONTEXT]\ncomponent: <name>block near the top of your kickoff message. When the verification harness invokes you for anagent-qastep that declares acomponent:field, it prepends this context block to your prompt. If present, prefer that component. - Else, if multiple components have
config.qa_start_command, use the component whosenamematches the project name. - Else use the first component with
config.qa_start_command.
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.
- 3d ago First seen · 288 lines · 16 tokens per session scan C d5859add1d5a
qa-test is a skill published in the GitHub repository G-Research/bobbit (11 stars, last pushed 3d ago), licensed MIT. It adds 16 tokens to every session and 3,425 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
use-agent-browser-for-airi
Test AIRI display-model imports with agent-browser across stage-tamagotchi Electron, stage-web, and stage-pocket mobile web layouts. Use when uploading and verifying contributor-supplied Live2D ZIP, VRM, or MMD ZIP/PMX/PMD files through AIRI's model selector, including onboarding bypass, format-specific import…
e2e-deployment-skill
Use this deployment skill to verify shared skills load during Playwright startup.
playwright-screen-recording
Record browser test videos with Playwright for PR review and bug fix verification.
actionbook-web-test
Run browser-based web tests against websites using Actionbook CLI. Activate when the user wants to test a website workflow, run smoke tests, verify a user flow, check if a web application works, run regression tests, or validate browser-based interactions. Supports test definition, execution, assertion, reporting, and…
web-preview
Flutter Web版をビルド → サーバー起動 → Playwright でアクセス確認 → URLをユーザーに案内する。.
qa/e2e-playwright
Playwright E2E 测试完整方法论，涵盖项目初始化、Page Object Model、认证复用、API Mock、视觉回归、多浏览器测试、CI 集成和调试技巧.