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 ikraamg/csstruth --skill csstruthgit clone --depth 1 https://github.com/ikraamg/csstruthWrote 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/ikraamg/csstruth/csstruth)<a href="https://agentmods.dev/skills/ikraamg/csstruth/csstruth"><img src="https://agentmods.dev/badge/skills/ikraamg/csstruth/csstruth/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/ikraamg/csstruth/csstruth"><img src="https://agentmods.dev/badge/skills/ikraamg/csstruth/csstruth.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 | $0.00000 | $0.02138 |
| Opus 5.5 | $0.00000 | $0.00855 |
| Sonnet 5.5 | $0.00000 | $0.00428 |
| Haiku 4.5 | $0.00000 | $0.00214 |
Grade A, and why
csstruth 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
csstruth: CSS ground truth for agents
csstruth extracts the browser's ACTUAL rendered layout as deterministic text.
Eleven tools — six inspectors, one composite verdict (verify), one baseline
writer (baseline), one load-time stability report (stability) — as MCP
tools (mcp__csstruth__*) or the csstruth CLI.
The one hard rule
Never claim anything about what renders without reading it from csstruth. No guessing from source, no "this should now be centered". Diagnose with the tool, edit the source, re-verify with the tool.
Workflow
1. Before touching any CSS on a working page — lock the good state:
snapshot with a name and an ABSOLUTE dir (e.g. <project>/.csstruth).
Relative dirs resolve against the MCP server's cwd, not the project. For
responsive work snapshot with viewports so diffs exist per breakpoint.
2. Diagnose (in this order):
layout— the whole rendered tree: positions(x,y WxH), flex/grid facts, inline⚠warnings. Scope withselector/depthon big pages.inspect— one element in depth: box model, non-default styles, stacking context, why it has its width/height.explain— REQUIRED before editing any rule you didn't write: which declaration wins (file:line, source-mapped), which lost and why, and what layout constraint overrides the declared value (flex-basis, min/max, grid). Fix the rule explain names — not the first plausible one in the source.
3. After every edit: diff against your snapshot. It reports exact moves/
resizes in px. (no layout changes) + your intended change = success. Unrelated
movement = your edit had side effects; explain them before proceeding.
Editing for a while? Start watch instead of re-running diff. CLI only
(csstruth watch <url>, no MCP tool — a streaming daemon doesn't fit
request/response): start it once in a background shell, then read its stream
as you edit instead of re-running diff/check after every change. It
prints the layout delta and any NEW/RESOLVED violations under a [HH:MM:SS]
block on each real change, and stays quiet otherwise — no need to re-invoke a
tool per edit. Stop it (Ctrl+C / kill the background job) when you're done.
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 Changed · +2 lines 445434964624
- 1mo ago First seen · 151 lines · 0 tokens per session scan A aba22c031528
csstruth is a skill published in the GitHub repository ikraamg/csstruth (2 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,138 tokens. 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
popular-web-designs
54 real design systems (Stripe, Linear, Vercel) as HTML/CSS.
video-shot-demos
A workflow for creating animated HTML pages, one page per video shot, lesson segment, or narrated scene. It combines a shared player structure with different visual designs for each shot.
designlang-tokens
Use when styling UI for cal.com — references the extracted design system tokens instead of inventing colors, spacing, or typography.
hero-slider-settle
A render-quality slider rises past the target, corrects, then settles on 75%. Function: range control / preference set.
animate
Build a web animation: whether it should animate at all, the purpose, the tool, the properties, the easing curve, the duration, interruption and exit. Writes the CSS or Motion code for dropdowns, modals, toasts, tooltips, accordions, press feedback, stagger and scroll reveals. Use to animate something or make a…
reimagine-it
Content-Derived Design — reads existing HTML and writes a stronger standalone page from headings, facts, names, dates, numbers, links, emails, and colors already in that file. CLI: npx reimagine-it --auto -i page.html -o out.html; also variations, lock, and audit. Use when the user says /reimagine-it, "reimagine it"…