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 agentmods add skills/g-research/bobbit/htmlnpx skills add G-Research/bobbit --skill htmlgit clone --depth 1 https://github.com/G-Research/bobbitWhat 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 | $0.00018 | $0.01642 |
| Opus 5 | $0.00009 | $0.00821 |
| Sonnet 5 | $0.00004 | $0.00328 |
| Haiku 4.5 | $0.00002 | $0.00164 |
Grade A, and why
html 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 2d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Interactive HTML UI
Build an interactive HTML interface in the live preview panel for: $ARGUMENTS
How it works
Use preview_open(html="...") to render ephemeral reports, mockups, charts, and interactive UI in a split-pane alongside the chat. The panel auto-updates on each call. The user can expand it fullscreen via the maximize icon (Escape to exit).
The preview workspace dedupes identical artifacts by content hash. Reopening or regenerating the same bytes may select/update the existing live preview tab instead of creating a second historical tab; different content remains separately restorable.
Use preview_open(file="/absolute/path/to/file.html") only for an existing or reusable file-backed preview. Declare sibling CSS, images, and other assets explicitly with assets or manifest; undeclared siblings will 404. Do not write and preview the same artifact through both the inline file-render surface and the live preview panel.
For the full contract, follow defaults/docs/html-rendering.md.
Setup — always start with this skeleton
<link rel="stylesheet" href="/src/ui/app.css">
<style>
/* Your custom styles here — use CSS variables from app.css.
NEVER alias a surface token with a single-mode fallback like
`--muted: var(--muted-foreground, #9aa0ad)`, and never put surface
tokens in your own :root{} — both freeze one mode and make muted text
INVISIBLE in standalone tabs / the bridge race. Reference
var(--muted-foreground) directly; the injected snapshot supplies it. */
body {
font-family: var(--font-sans, system-ui, sans-serif);
background: var(--background);
color: var(--foreground);
margin: 0;
padding: 16px;
}
</style>
<!-- Your UI here -->
<script>
// Your interactivity here
</script>
The <link> gives you the full Bobbit design system: Tailwind CSS 4, theme variables, and component styles. The preview iframe is same-origin with the Vite dev server, so this works out of the box.
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.
- 2d ago First seen · 149 lines · 18 tokens per session scan A 4f4af136eae5
html is a skill published in the GitHub repository G-Research/bobbit (11 stars, last pushed 2d ago), licensed MIT. It adds 18 tokens to every session and 1,642 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
use-agent-browser-for-airi
Test AIRI display-model imports with agent-browser across stage-tamagotchi Electron, stage-web, and stage-pocket mobile web layouts. Use when uploading and verifying contributor-supplied Live2D ZIP, VRM, or MMD ZIP/PMX/PMD files through AIRI's model selector, including onboarding bypass, format-specific import…
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
browser_cdp
通过 Chrome DevTools Protocol (CDP) 连接已运行的浏览器，或扫描本机 CDP 端口，用于远程调试与多工具共享浏览器实例。.