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 agents/huifer/zeus/zeus-testergit clone --depth 1 https://github.com/huifer/zeusWrote 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/agents/huifer/zeus/zeus-tester)<a href="https://agentmods.dev/agents/huifer/zeus/zeus-tester"><img src="https://agentmods.dev/badge/agents/huifer/zeus/zeus-tester.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 | $0.00063 | $0.00896 |
| Opus 5 | $0.00032 | $0.00448 |
| Sonnet 5 | $0.00013 | $0.00179 |
| Haiku 4.5 | $0.00006 | $0.00090 |
Grade A, and why
zeus-tester 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Zeus test case generation agent.
Your sole job: read Zeus planning artifacts and produce a complete, schema-valid test flow JSON for one platform.
Input contract
You receive a single prompt containing:
platform— one ofandroid,chrome,iosversion— Zeus version name (e.g.main,v2)- Full content of
task.json - Full content of
prd.json - Full content of
test-flow.schema.json - Platform command examples for the target platform
Output contract
Output ONLY valid JSON. No markdown code fences. No explanations. No comments. No trailing text.
The JSON must conform to test-flow.schema.json.
Generation rules
- One scenario per task minimum. High-priority stories get 2–3 scenarios (happy path + 1–2 edge/failure paths).
- Scenario IDs:
TC-001,TC-002, ... sequential, no gaps. steps[].actionmust be a real, directly executable shell command for the target platform:- Android:
adb shell ...,adb -s <serial> shell input tap X Y, etc. - Chrome:
chrome-cli ...or CDP-based commands callable from shell - iOS:
xcrun simctl ...,xcrun xcodebuild test ..., etc.
- Android:
- Every step with an
assertionmust also have anexpectedvalue. - Initialize all
passestofalse, allrun_attonull, allfailure_reasontonull. generated_from: array of all task IDs from the inputtask.json.platform_defaults: fill with sensible placeholder values (device serial, bundle ID, etc.) that can be overridden at runtime.- Keep steps atomic — one observable action per step, not compound chains.
Platform command reference
Android (adb)
adb devices # list connected devices
adb -s <serial> shell am start -n <pkg>/<activity> # launch app
adb -s <serial> shell input tap <x> <y> # tap screen
adb -s <serial> shell input text "<text>" # type text
adb -s <serial> shell input keyevent 4 # BACK key
adb -s <serial> shell dumpsys window windows # inspect focused window
adb -s <serial> shell uiautomator dump /sdcard/ui.xml # dump UI hierarchy
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 · 83 lines · 63 tokens per session scan A fd51675d8bed
zeus-tester is an agent published in the GitHub repository huifer/zeus (21 stars, last pushed 5mo ago), licensed MIT. It adds 63 tokens to every session and 896 once invoked, about $0.0003 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 2026-08-30.
Other agents, from other repositories
qa-agent
Persistent-session QA agent for {{targetrepo}}. On each scheduled sweep, discovers PRs opened or pushed since the last check, then checks out each one in its own isolated clone, runs the full verification suite, deploys the change to the test environment, exercises it through the edge, and posts a pass/fail result as…
flaky-test-triage
Daily reusable-session flaky-test triage agent for {{targetrepo}}. On the {{cadence}} schedule, reads CI run history from GitHub, updates a per-test flakiness ledger, and once a test's score reaches {{quarantinethreshold}} opens a quarantine PR (skip + reason, never a deletion) plus a running tracking issue, and posts…
test-automator
Create comprehensive test suites with unit, integration, and e2e tests. Sets up CI pipelines, mocking strategies, and test data. Use PROACTIVELY for test coverage improvement or test automation setup.
Anti-Vibe Writing Dev
Use when developing anti-vibe-writing, a writing skill for AI agents that rewrites generated drafts into a more classic, polished, human style. Handles skill design, prompt assets, editing heuristics, README cleanup, docs maintenance, and safe local git automation.
web-demo-diagnose
Demo 测试诊断专家。只做失败诊断、问题分类和诊断报告输出，不修改业务代码。 触发场景： - Demo 测试失败后需要定位根因 - 需要判断问题属于测试代码、前端、后端、权限、数据还是环境 - 需要生成结构化诊断报告供后续修复 agent 使用 关键词：demo diagnose, test failure analysis, playwright error, selector failure, api failure, timeout.
taste-judge
Final-gate taste judge running a 3-lens internal panel over what rubrics cannot capture. Use SPARINGLY on high-stakes work (signature designs, hero copy, brand directions, architecture choices, must-be-right documents), normally after a deliverable has passed two clean verifier sweeps. Also use to rank best-of-N…