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 unshift-ai/unshift-mcp --skill unshift-website-buildergit clone --depth 1 https://github.com/unshift-ai/unshift-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/unshift-ai/unshift-mcp/unshift-website-builder)<a href="https://agentmods.dev/skills/unshift-ai/unshift-mcp/unshift-website-builder"><img src="https://agentmods.dev/badge/skills/unshift-ai/unshift-mcp/unshift-website-builder/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/unshift-ai/unshift-mcp/unshift-website-builder"><img src="https://agentmods.dev/badge/skills/unshift-ai/unshift-mcp/unshift-website-builder.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 | $0.00083 | $0.01395 |
| Opus 5.5 | $0.00033 | $0.00558 |
| Sonnet 5.5 | $0.00017 | $0.00279 |
| Haiku 4.5 | $0.00008 | $0.00139 |
Grade A, and why
unshift-website-builder 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 2mo 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Unshift website builder
Unshift turns this conversation into a real, published website. You drive it through the Unshift MCP connector: you describe and author the site with the connector's tools, and Unshift stores, renders, and publishes it to a live URL the user owns. The user's own account runs the build; everyday edits later happen in Unshift's visual editor.
Use this skill whenever the user wants to build a new site, rebuild or import an existing site from a URL, or edit and publish an Unshift site.
Golden rules
- Collaborate; never build a whole site in one pass. Ask 2 to 4 focused questions first (business, audience, pages, vibe, must-have sections), unless the user gave a screenshot or a very detailed brief. Confirm the direction, then build incrementally and show progress.
- Call
block_guidebefore you author or edit any block. It returns the component shape and the typed prop-value contract. Re-read it whenever you are unsure. - Typed prop values are mandatory. A link is always a
LinkTypeobject ({ id: "", href: "/path", target: "_self" }), never a bare string or aurlfield. Images areImageType, icons areIconType, forms areFormType. Untyped values silently fail to resolve. - Finish every build with
finalize_app. It validates, saves, and (on a brand-new site's first finalize) publishes automatically, returning both a livesiteUrland aneditorUrl. The site is not done until you call it.
Build a site from scratch
create_appto create the app (returnsappId,appUrl). This is the first step; commit to the whole sequence.generate_themeimmediately after, with a complete palette: light and dark values for all 19 color tokens, plus radius, heading font, and body font. PassaiContext(the design brief) so later sessions can reuse it.set_layoutto place the shared header and footer. Author them as blocks with a semantic<header>/<footer>root element (not a<Section>wrapper). The layout shows on every page, so never put the header or footer in page content.add_pagefor each page. Use slug""for the home page. Author each block as{ componentName, content }perblock_guide, or reuse one of this app's existing blocks with{ blockId, props }to repeat a section across pages instead of duplicating it. Passseoand anyassetMap.finalize_appto publish. Share both thesiteUrl(live in about 1 to 2 minutes) and theeditorUrl.
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.
- 2mo ago First seen · 54 lines · 83 tokens per session scan A 658e243969b9
unshift-website-builder is a skill published in the GitHub repository unshift-ai/unshift-mcp (0 stars, last pushed 4mo ago), licensed MIT. It adds 83 tokens to every session and 1,395 once invoked, about $0.0003 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.
unshift-website-builder compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| unshift-website-builderunshift-ai/unshift-mcp | Skill | 0 | 4mo ago | MIT | A | 83 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | yesterday | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 100k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 314k | 4d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |