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.
git clone --depth 1 https://github.com/smk-labs/claude-pluginsWrote 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/plugins/smk-labs/claude-plugins/readable)<a href="https://agentmods.dev/plugins/smk-labs/claude-plugins/readable"><img src="https://agentmods.dev/badge/plugins/smk-labs/claude-plugins/readable/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/plugins/smk-labs/claude-plugins/readable"><img src="https://agentmods.dev/badge/plugins/smk-labs/claude-plugins/readable.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
readable 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 19d 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 — 21 lines — stays where its author put it; the contents beside it link to each section on GitHub.
{
"name": "readable",
"description": "Beautiful replies as styled cards (Persian/RTL always, English by default): the model sends content only, the template carries all styling. KPI grids, charts, trend sparklines, timelines, flows, callouts, badges, CTA buttons, content cards, side-by-side columns, quotes, section numbering and section breaks, a minimal icon set, link previews with an optional live frame of the target page, a pinned tab bar for a long report's sections, and screenshots or looping /fig animations inlined so a report stays one offline file. The kit is pay-per-use: the template carries only the base sheet and each card is served just the component CSS it uses. Delivery picks itself from the tools a host actually has: the MCP Apps card server where one is connected, an inline widget kit where the host has a widget tool, and BiDi-safe plain text where nothing can render. Only the desktop-registered server can ever offer the card tool; the plugin's own scoped server carries the export tools alone, under a different name, so it can never shadow the one that paints. Installing it is the whole setup: the first session registers the desktop card server in every Claude profile on the machine, at a version-free path an update cannot break, writes one backup per config and never touches them again, and says so once. /readable:connect disconnect opts out everywhere and sticks. Cards can also render straight from *-card.html files written by background workers (htmlFile mode, zero context cost). Projects can carry a committable .readable/ brand layer (brand skill) that reskins their cards and reports. Every artifact closes with one muted signature line, defined once in the kit and injected at build time, which a project can switch off. Plus report export and a visualize skill.",
"version": "7.0.1",
"author": {
"name": "Seyed M. Keramati",
"url": "https://github.com/SMKeramati"
},
"mcpServers": {
"readable-local": {
"command": "node",What it installs
The manifest is a name and a version. 4 skills, 1 hook travel with it, and installing the plugin installs all of them — 234 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 19d ago Changed dbb9c0271b3a
- 21d ago Changed 42ba0f5ffb0a
- 1mo ago Changed 6b352c1971a6
- 1mo ago Changed 6b36c530d906
- 1mo ago First seen · 21 lines scan A bfefb00c136e
readable is a plugin published in the GitHub repository smk-labs/claude-plugins (11 stars, last pushed 20d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 plugins, from other repositories
handbook-visual-explainer
Generate beautiful HTML pages for diagrams, diff reviews, plan reviews, slides, and data tables.
bash-specialist
Production-quality bash scripting with shellcheck compliance, robust error handling, and beautiful terminal UX via Charmbracelet gum. Skills: gum-beautify (TTY-safe interactive output). MCP: context7.
website-audit
Website SEO/GEO, design research, usability, and WCAG accessibility audit skills.
site-forge
Build a distinctive, production-grade website (and optional Remotion promo video) for any business, product, or person, then deploy it free. Two modes: auto (fast, autonomous) and full (best quality).
brand-forge
Generate on-brand logos, social templates, and marketing graphics for any brand — vector assets need no accounts or keys; AI imagery is opt-in.
design-thinking
Design Thinking PRD Generator — guides users from raw pain points through Empathize, Define, Ideate, Prototype, and PRD compilation to produce vbounce-compatible Product Requirements Documents.