Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/squidbay/factorynpx agentmods add skills/squidbay/factory/factory-render-verifyWrote 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/squidbay/factory/factory-render-verify)<a href="https://agentmods.dev/skills/squidbay/factory/factory-render-verify"><img src="https://agentmods.dev/badge/skills/squidbay/factory/factory-render-verify/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/squidbay/factory/factory-render-verify"><img src="https://agentmods.dev/badge/skills/squidbay/factory/factory-render-verify.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 | $0.00133 | $0.01569 |
| Opus 5.5 | $0.00053 | $0.00628 |
| Sonnet 5.5 | $0.00027 | $0.00314 |
| Haiku 4.5 | $0.00013 | $0.00157 |
Grade A, and why
factory-render-verify 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 1mo 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
factory-render-verify — receipts, not vibes
A page can look right in one browser window and fall apart on a phone, in landscape, or on a
wide monitor — and you would never know from reading the code. This skill renders the real page
at every screen size, measures the things that actually break, and writes the results to disk as
receipts: a screenshot per size plus a report.json of hard numbers. Code attaches those to
the pull request; Team Leader audits against them; you merge. Receipts are the gate, not the claim.
This is the render half of the design quality gate — guides/DESIGN-QUALITY-GATE.md is the contract that says what "done" means; this skill is how STEP 7's visual QA gets proven instead of asserted.
What it measures, at every size
- Horizontal overflow — the page wider than the screen. The single most common mobile bug, and invisible in a desktop window.
- Headline and body type size — catches an oversized hero headline and, worse, body copy under 16px (too small to read comfortably on a phone).
- Tap targets under 44px — buttons and links smaller than a fingertip.
- Safe-area handling — whether the page accounts for the notch and the home indicator on modern phones.
- Monospace on body copy — the monospaced font belongs on labels and numbers, never on paragraphs; it's a tell that a page was styled by feel rather than by system.
- Zoom sanity — records that the render was at true 1:1 (see the zoom-check rule below).
The screen sizes it renders
Phone portrait 390×844 and 430×932 · phone landscape 844×390 (phones turn sideways —
a hero built only for portrait collapses here) · tablet 834×1112 · desktop 1280×720,
1440×900, 1920×1080. These match the widths in the quality gate's STEP 7. Override with
--sizes "name:WxH,..." for a page with unusual breakpoints.
Run it
The browser ships with your session — nothing to download. Install the Python binding once if
it's missing (do not run playwright install; the browser is already on disk):
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.
- 1mo ago First seen · 111 lines · 133 tokens per session scan A fbceb17519ae
factory-render-verify is a skill published in the GitHub repository squidbay/factory (5 stars, last pushed 2mo ago), licensed MIT. It adds 133 tokens to every session and 1,569 once invoked, about $0.0005 per session on Opus 5.5. 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
product-test
Use this skill to define the observable behavior that must be true for a story or feature to be considered done. It turns product context into concise, testable Given/When/Then scenarios that engineers and QA can verify without guessing intent.
verify
Exercise the real app/API/CLI and collect observable evidence; tests alone do not count as end-to-end verification.
webapp-testing
Start/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.
ui-craft-checks
Use this comprehensive gate for formal review, QA, launch handoff, exact-fidelity inspection, high-risk complex UI, or when the fast gate exposes a deeper craft problem. Ordinary single-screen work uses ui-design-executor and its bundled validator without loading this full matrix. Pair standalone artifacts with…
ui-html-renderer
Use this specialist for unusually complex stateful HTML, blank/runtime-risk repair, strict source-to-HTML fidelity, reusable component galleries, brand boards, or detailed implementation handoff. Ordinary single-page HTML uses ui-design-executor, which already preserves HTML-default delivery and baseline…
qa
Browser-based QA verification. Launches a real browser, navigates the app, clicks buttons, fills forms, and tests user flows. Works as a standalone skill or as a phase end condition in campaigns. Requires Playwright (optional dependency, graceful skip if not installed).