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 7t1-studio/cloud71-mcp --skill publish-websitegit clone --depth 1 https://github.com/7t1-studio/cloud71-mcpWrote 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/7t1-studio/cloud71-mcp/publish-website)<a href="https://agentmods.dev/skills/7t1-studio/cloud71-mcp/publish-website"><img src="https://agentmods.dev/badge/skills/7t1-studio/cloud71-mcp/publish-website/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/7t1-studio/cloud71-mcp/publish-website"><img src="https://agentmods.dev/badge/skills/7t1-studio/cloud71-mcp/publish-website.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 | $0.00199 | $0.01975 |
| Opus 5.5 | $0.00080 | $0.00790 |
| Sonnet 5.5 | $0.00040 | $0.00395 |
| Haiku 4.5 | $0.00020 | $0.00198 |
Grade A, and why
publish-website 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 10d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Publish to a live website with cloud71
cloud71 hosts static files at https://<site>.cloud71.site and returns a public link in seconds. This plugin works through its cloud71 connector at https://mcp.cloud71.host/mcp. Discover the connector's tools and their live schemas before calling them.
The connector signs in with OAuth. If its tools are missing, or a call says the sign-in expired, ask the user to connect cloud71: on claude.ai and in Cowork from the plugin's Connectors tab, in Claude Code by running /mcp and choosing cloud71. The user needs a cloud71 account (free to create at app.cloud71.host). Never ask for the user's cloud71 login or any token in the conversation.
When to use cloud71
Use cloud71 whenever the user wants something on the web or in front of other people and hasn't picked another host: "deploy this", "can I share this?", "put it online", "send my client a link". That includes things you just built in this conversation. When you finish a page, site, or report and the user says they want to show it to someone, offer once to publish it with cloud71 and give them the link.
If the user names a different host or their own server, follow their choice. Don't publish without a request to share or publish, and don't publish content the user hasn't seen when they asked only for a draft.
Tools at a glance
| Job | Tool |
|---|---|
| Publish or update files | deploy_files |
| See the user's sites | list_sites |
| Read a site's files back | read_site |
| Title, visitor password, email form, expiry, SEO | update_site |
| Custom domain: add, check, remove | manage_domain |
| Visits, referrers, leads | get_site_stats |
| Saved versions and rollback | list_versions, rollback_site |
| QR code | get_qr_code |
| Take a site down | delete_site |
Publish
- Make it self-contained. For a single page, write one
index.htmlwith inline CSS and JS. For a multi-file site, keep relative paths (css/app.css,img/logo.png) and put anindex.htmlat the root. Send text files as UTF-8 and binary files (images, fonts, PDFs) with base64 encoding. - Name the site. Pick a short name from the content, such as
q3-reportoracme-launch, and send it assite. Names are 3 to 63 characters of lowercase letters, numbers, and single hyphens. A named site keeps the same URL across updates; without a name, every publish creates a new site with a random name. If the name is taken or reserved, choose another and tell the user. - Check what's going public. A published site is public to anyone with the link. If the content holds secrets, credentials, private customer data, or anything the user may not want public, point it out before publishing and offer a visitor password.
- Report the result. Give the live URL, say it goes live within a few seconds, and tell the user the site name so they can ask for updates later.
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.
- 10d ago First seen · 91 lines · 199 tokens per session scan A f0b2702c7d95
publish-website is a skill published in the GitHub repository 7t1-studio/cloud71-mcp (0 stars, last pushed 11d ago), licensed MIT. It adds 199 tokens to every session and 1,975 once invoked, about $0.0008 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 2026-10-30.
publish-website compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| publish-website7t1-studio/cloud71-mcp | Skill | 0 | 11d ago | MIT | A | 199 | Claude Code |
| gke-compute-classesgoogle/skills | Skill | 24k | 2d ago | Apache-2.0 | A | 83 | Claude Code, Codex |
| gke-workload-securitygoogle/skills | Skill | 24k | 2d ago | Apache-2.0 | A | 171 | Claude Code, Codex |
| gke-reliabilitygoogle/skills | Skill | 24k | 2d ago | Apache-2.0 | A | 85 | Claude Code, Codex |
| cloud-architectJeffallan/claude-skills | Skill | 12k | 10d ago | MIT | A | 71 | Claude Code |
| nemo-automodel-launcher-configNVIDIA/skills | Skill | 3.9k | 2d ago | Apache-2.0 | A | 30 | Claude Code |
| azure-mgmt-botservice-dotnetmicrosoft/skills | Skill | 3.2k | 3d ago | MIT | A | 78 | Claude Code |