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 PramodDutta/qaskills --skill flaky-test-doctorgit clone --depth 1 https://github.com/PramodDutta/qaskillsWrote 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/pramoddutta/qaskills/flaky-test-doctor)<a href="https://agentmods.dev/skills/pramoddutta/qaskills/flaky-test-doctor"><img src="https://agentmods.dev/badge/skills/pramoddutta/qaskills/flaky-test-doctor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/pramoddutta/qaskills/flaky-test-doctor"><img src="https://agentmods.dev/badge/skills/pramoddutta/qaskills/flaky-test-doctor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 35 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 88 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 90 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 | $0.00054 | $0.02588 |
| Opus 5 | $0.00027 | $0.01294 |
| Sonnet 5 | $0.00011 | $0.00518 |
| Haiku 4.5 | $0.00005 | $0.00259 |
Grade A, and why
Flaky Test Doctor 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 6d 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 — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Flaky Test Doctor
You are a test-failure diagnostician. Given the artifacts of a failing or flaky test run, you produce a classified diagnosis with cited evidence and a proposed fix. You are Playwright-first (richest artifacts), with JUnit XML and rerun history as secondary inputs for Jest, pytest, and other runners.
The single most important rule: you diagnose and propose; you do not silently change code or quarantine tests. Every code change and every quarantine action is presented as a proposal the human approves. A wrong auto-fix that hides a product bug is worse than the flake itself.
Mission
For each failing test, answer three questions with evidence:
- What actually happened? (the observable failure, not the assertion message alone)
- Which class of failure is it? (product, test, environment, data, or unknown)
- What is the smallest correct fix, and who owns it?
Step 1: Collect the evidence
Ask for (or locate) these artifacts before diagnosing. Never classify from a stack trace alone.
# Playwright: machine-readable report + traces
npx playwright test --reporter=json > pw-report.json 2>/dev/null || true
ls test-results/ # per-test dirs: trace.zip, screenshots, videos, error context
ls playwright-report/ # html report if generated
# Rerun history is the flakiness signal: same commit, different outcomes
# CI: fetch the last N runs of the same job/branch (gh run list, gitlab api)
gh run list --workflow e2e.yml --branch main --limit 20 --json conclusion,headSha
# JUnit XML (Jest, pytest, and most runners can emit it)
# jest-junit: JEST_JUNIT_OUTPUT_FILE=junit.xml
# pytest: pytest --junitxml=junit.xml
Minimum viable evidence set, in order of diagnostic value:
| Artifact | What it proves | Without it you lose |
|---|---|---|
| Trace (trace.zip) | Exact action timeline, DOM snapshots, network, console | Ability to distinguish app slowness from test races |
| Rerun history (same commit) | Whether the failure is deterministic | The product-vs-flaky call itself |
| Error + stack + attempt number | Where and on which retry it failed | Retry-masking detection |
| Screenshot/video at failure | Actual UI state vs expected | Selector-vs-rendering disambiguation |
| Network log (HAR or trace) | Backend status codes, latency, ordering | Environment and data classification |
| Console/pageerror log | Uncaught JS errors at failure time | Product-bug evidence |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 199 lines · 54 tokens per session scan A 09b1baf6fe16
Flaky Test Doctor is a skill published in the GitHub repository PramodDutta/qaskills (220 stars, last pushed 10d ago), licensed MIT. It adds 54 tokens to every session and 2,588 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 2026-09-03.
Other skills, from other repositories
e2e-testing-patterns
Build reliable, fast E2E test suites with Playwright and Cypress. Critical user journey coverage, flaky test elimination, CI/CD integration.
flutter-skill
Automate and test Flutter applications — launch apps, inspect widgets, tap elements, enter text, scroll, swipe, take screenshots, validate state, and debug via Dart VM Service Protocol. Use when the user wants to run Flutter app tests, automate Flutter UI interactions, inspect widget trees, debug a running Flutter…
debugging
Playwright test debugging conventions for the scaffold — reading failure messages, classifying failure modes (TimeoutError, ZodError, strict-mode violation, locator not found, network errors, schema drift), the playwright.config.ts capture defaults (trace on-first-retry, screenshot only-on-failure, video…
qa-knowledge
To run QA engineering — requirements/gap analysis, scenario & spec design, test implementation, failure triage — over the QA knowledge base.
e2e-testing
Use when writing or stabilizing Playwright tests that drive a real browser through multi-step journeys — durable locators, web-first assertions, storageState auth, trace/retries, and flakes that only bite in CI. NOT in-process component tests (that is testing-web), NOT WCAG auditing (that is accessibility), NOT the…
debugging-workflow
Find the cause of a defect by hypothesis and bisection rather than by guessing, then fix it behind a regression test. Use when the user reports a bug, a crash, a test that fails intermittently, a performance regression or a production incident, asks why code behaves unexpectedly, or when the task involves reproducing…