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 CassioRoos/godfly-skills --skill mean-qagit clone --depth 1 https://github.com/CassioRoos/godfly-skillsWrote 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/cassioroos/godfly-skills/mean-qa)<a href="https://agentmods.dev/skills/cassioroos/godfly-skills/mean-qa"><img src="https://agentmods.dev/badge/skills/cassioroos/godfly-skills/mean-qa/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/cassioroos/godfly-skills/mean-qa"><img src="https://agentmods.dev/badge/skills/cassioroos/godfly-skills/mean-qa.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00062 | $0.01320 |
| Opus 5 | $0.00031 | $0.00660 |
| Sonnet 5 | $0.00012 | $0.00264 |
| Haiku 4.5 | $0.00006 | $0.00132 |
Grade A, and why
mean-qa 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 today.
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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MeanQA
Find defects a happy-path pass misses. Preserve what was actually tested, including passing, failed, blocked, and interrupted work. Never confuse a test plan with a run or a screenshot with working behavior.
First action: create the proof
Every campaign gets <project-root>/.proof/<timestamp-slug>/PROOF.md,
including planning-only and blocked campaigns. Before HTTP/browser/database
probes or test execution, run:
sh <skill-dir>/scripts/start-proof.sh <project-root> <run-slug> <environment> <build>
Derive the project root from the app/service under test; honor a user-specified
artifact location. Use unknown for metadata not yet established. Read back
the returned document before any probe. The helper never overwrites an earlier
run and publishes a nonempty initial document. If you are only designing,
immediately mark it DESIGNED — NOT RUN; if blocked, record BLOCKED,
the failed prerequisite and next action. No empty "successful" report.
If creation fails, stop probes, report the refused path and use another
user-permitted persistent location if available. Never pretend the file exists.
This is deterministic when invoked, not a runtime hook: skill instructions
cannot guarantee capture if the agent bypasses them or is killed. An abandoned
IN PROGRESS report means incomplete, never pass.
Update the report and case artifacts after each meaningful case, before the next
one. Save the pre-change state before changing it. On normal completion record
the actual end time, verdict, coverage and remaining gaps; on interruption,
record INCOMPLETE if execution still permits a write. Do not overwrite earlier
runs or silently remove evidence. Keep .proof local/private; check ignore rules
before staging and never automatically commit its contents.
Load only what the task needs
- Always read findings.md: oracle, false-positive control and honest verdicts.
- Planning: read attack.md and cases.md. Mark every proposed case not run.
- Execution: read environment.md and evidence.md before probing. Add attack/cases when designing new cases, not when simply re-running existing ones.
- Report formatting: read report.md when writing the completed report.
- External publishing only: read publishing.md once publishing is requested or already authorized. A PR existing, or a request to verify a PR, does not authorize uploads, comments or body edits.
What ships with it
9 files 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.
- today Changed · -163 lines · -16 tokens per session 1f479129eb4b
- 5d ago Changed · -3 lines 0b2e2196a232
- 9d ago First seen · 284 lines · 78 tokens per session scan A e7d3d293e3e1
mean-qa is a skill published in the GitHub repository CassioRoos/godfly-skills (1 stars, last pushed today), licensed MIT. It adds 62 tokens to every session and 1,320 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 2026-08-31.
Other skills, from other repositories
auto-test-code
A structured process for critically reviewing and testing software code. It records review findings, test plans, commands, results, and supporting files in a project workspace.
test-conversion-reviewer
Reviewer for converting unit tests to browser tests for Project Bedrock. Invoke this when the user needs to review their work while removing complex Browser dependencies from tests.
chrome-cdp
Drive a headless Chrome over the Chrome DevTools Protocol (CDP) for browser QA — navigate, click, fill forms, read the DOM/accessibility tree, screenshot, and assert. Use whenever a task requires loading a web page and interacting with it like a user. Chrome is launched by a bash step (recipe below); this skill…
agent-browser
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
remove-ai-slops
Removes AI-generated code smells from branch changes or an explicit file list behind regression tests. Use when the user asks to clean up, deslop, or remove AI-slop patterns from recent changes.
codex-loop
Fix Elixir/Phoenix code until Codex CLI review comes back clean — bounded review, fix, verify loop before opening a PR. Use when codex is installed and you want an external cross-model critic on your changes before pushing.