Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Llm Response Beautifier

    1

    Turns a long AI answer into one self-contained HTML page: headings, contents, styled tables, callouts and code, with a light/dark switch and no JavaScript

    $5

    /mo

    Subscription · cancel anytime

    Secure checkout via Stripe

    • Always the latest version
    • Delivered live to your agent via MCP
    • Cancel anytime, access ends at period end
    Security scanned
    Llm Response Beautifier

    Llm Response Beautifier

    Example session with this skill installed

    Prettify this response as HTML

    Choosing a vector database

    Short answer: start with pgvector if you already run Postgres; move to a dedicated store only when recall or latency forces you to.

    Options compared

    OptionBest forWatch out for
    pgvectorteams already on Postgresindex build time above ~10M rows
    Qdrantfiltered search at scaleanother service to operate
    Pineconezero opscost and vendor lock-in

    Steps

    1. Measure your corpus size and query rate.
    2. Prototype with pgvector and an HNSW index.
    3. Benchmark recall@10 against a held-out set.

    Note: embeddings from different models are not interchangeable - re-embed when you switch.

    CREATE INDEX ON items USING hnsw (embedding vector_cosine_ops);

    • Read your context and instructions
    • Compiled the llm response beautifier
    • Generated the UI component

    Target: the response you pasted into your message ("Choosing a vector database"), not my previous reply.

    This environment cannot render a live inline widget, so I saved the page as response.html. Open it in any browser to view it. It opens in dark mode, and the pill at the top switches to light mode with no JavaScript. The full HTML is below.

    <!DOCTYPE html>
    <html lang="en">
    <head>
    <meta charset="utf-8">
    <meta name="viewport" content="width=device-width, initial-scale=1">
    <style>
      :root {
        --bg: #0f1320; --bg-soft: #171c2e; --card: #1c2238; --text: #e8ecf8;
        --muted: #9aa6c8; --border: #2a3252; --accent: #7c8cff; --accent-2: #56d4c4;
        --code-bg: #0b0f1c; --code-text: #d6e0ff; --quote: #b8c2e6; --link: #8ab4ff;
        --info: #5b8cff; --success: #3fce9a; --warning: #f2b34b; --danger: #f2657a;
        --row-alt: #20273f; --chart-grid: #2a3252;
      }
      #theme:checked ~ * {
        --bg: #f7f8fc; --bg-soft: #eef1fb; --card: #ffffff; --text: #1b2138;
        --muted: #5a6488; --border: #e2e6f4; --accent: #5b6bff; --accent-2: #1eb8a6;
        --code-bg: #0d1117; --code-text: #e6edf3; --quote: #4a5478; --link: #2563eb;
        --info: #2563eb; --success: #0f9d6b; --warning: #b9760c; --danger: #d23b54;
        --row-alt: #f3f5fd; --chart-grid: #e2e6f4;
      }
      #theme { display: none; }
      * { box-sizing: border-box; }
      body {
        margin: 0; padding: 2.5rem 1rem; min-height: 100vh;
        font: 16px/1.7 -apple-system, system-ui, "Segoe UI", sans-serif;
        color: var(--text);
        background:
          radial-gradient(900px 500px at 12% -8%, color-mix(in srgb, var(--accent) 22%, transparent), transparent 60%),
          radial-gradient(800px 520px at 100% 0%, color-mix(in srgb, var(--accent-2) 18%, transparent), transparent 55%),
          var(--bg);
        transition: background .35s ease, color .35s ease;
      }
      main {
        max-width: 820px; margin: 0 auto; background: var(--card);
        border: 1px solid var(--border); border-radius: 18px;
        padding: 2.25rem 2.5rem; box-shadow: 0 24px 60px -28px rgba(0,0,0,.55);
      }
      h1, h2, h3 { line-height: 1.25; letter-spacing: -.01em; margin-top: 1.8rem; scroll-margin-top: 1rem; }
      h1 {
        font-size: 1.9rem;
        background: linear-gradient(90deg, var(--accent), var(--accent-2));
        -webkit-background-clip: text; background-clip: text; color: transparent;
      }
      h2 { color: var(--text); border-bottom: 1px solid var(--border); padding-bottom: .35rem; }
      h3 { color: var(--accent); }
      a { color: var(--link); text-decoration: none; border-bottom: 1px solid color-mix(in srgb, var(--link) 40%, transparent); }
      a:hover { border-bottom-color: var(--link); }
      pre { background: var(--code-bg); color: var(--code-text); padding: 1.1rem 1.25rem; border-radius: 12px; overflow-x: auto; border: 1px solid var(--border); }
      code { font-family: ui-monospace, "SF Mono", Menlo, monospace; font-size: .9em; }
      p code, li code, td code { background: var(--bg-soft); padding: .12em .4em; border-radius: 6px; color: var(--accent-2); }
      blockquote { border-left: 3px solid var(--accent); margin: 1.2rem 0; padding: .4rem 1rem; color: var(--quote); background: var(--bg-soft); border-radius: 0 10px 10px 0; }
      ul, ol { padding-left: 1.3rem; }
      li::marker { color: var(--accent); }
      details { border: 1px solid var(--border); border-radius: 12px; padding: .6rem 1.1rem; margin: 1.1rem 0; background: var(--bg-soft); }
      summary { cursor: pointer; font-weight: 600; color: var(--accent); }
      .t-scroll { overflow-x: auto; border: 1px solid var(--border); border-radius: 12px; margin: 1.3rem 0; }
      table.t { width: 100%; border-collapse: collapse; font-size: .94rem; }
      table.t thead th { position: sticky; top: 0; background: var(--bg-soft); color: var(--text); text-align: left; padding: .7rem .9rem; border-bottom: 2px solid var(--border); }
      table.t td { padding: .65rem .9rem; border-bottom: 1px solid var(--border); }
      table.t tbody tr:nth-child(even) { background: var(--row-alt); }
      table.t tbody tr:hover { background: color-mix(in srgb, var(--accent) 12%, transparent); }
      .callout { display: flex; gap: .7rem; align-items: flex-start; border: 1px solid var(--border); border-left: 4px solid var(--accent); border-radius: 10px; padding: .8rem 1rem; margin: 1.2rem 0; background: var(--bg-soft); }
      .callout .ic { font-size: 1.1rem; line-height: 1.5; }
      .callout-info { border-left-color: var(--info); }
      .callout-success { border-left-color: var(--success); }
      .callout-warning { border-left-color: var(--warning); }
      .callout-danger { border-left-color: var(--danger); }
      .badge { display: inline-block; font-size: .76rem; font-weight: 600; padding: .15em .6em; border-radius: 999px; border: 1px solid var(--border); background: var(--bg-soft); color: var(--muted); }
      .badge-info { color: var(--info); border-color: color-mix(in srgb, var(--info) 45%, transparent); }
      .badge-success { color: var(--success); border-color: color-mix(in srgb, var(--success) 45%, transparent); }
      .badge-warning { color: var(--warning); border-color: color-mix(in srgb, var(--warning) 45%, transparent); }
      .badge-danger { color: var(--danger); border-color: color-mix(in srgb, var(--danger) 45%, transparent); }
      .kpi-grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(150px, 1fr)); gap: .9rem; margin: 1.3rem 0; }
      .kpi { background: var(--bg-soft); border: 1px solid var(--border); border-radius: 14px; padding: 1rem 1.1rem; }
      .kpi .label { font-size: .8rem; color: var(--muted); }
      .kpi .value { font-size: 1.7rem; font-weight: 700; margin-top: .2rem; background: linear-gradient(90deg, var(--accent), var(--accent-2)); -webkit-background-clip: text; background-clip: text; color: transparent; }
      .kpi-delta { font-size: .8rem; font-weight: 600; }
      .kpi-delta.up { color: var(--success); }
      .kpi-delta.down { color: var(--danger); }
      figure.chart { margin: 1.4rem 0; }
      figure.chart svg { width: 100%; height: auto; }
      figure.chart figcaption { font-size: .82rem; color: var(--muted); margin-top: .4rem; text-align: center; }
      nav.toc { background: var(--bg-soft); border: 1px solid var(--border); border-radius: 14px; padding: 1rem 1.2rem; margin: 1.5rem 0; }
      nav.toc strong { display: block; font-size: .8rem; text-transform: uppercase; letter-spacing: .06em; color: var(--muted); margin-bottom: .5rem; }
      nav.toc ol { margin: 0; }
      ol.footnotes { margin-top: 2.5rem; padding-top: 1rem; border-top: 1px solid var(--border); font-size: .85rem; color: var(--muted); }
      .theme-switch { display: inline-flex; align-items: center; gap: .5rem; cursor: pointer; user-select: none; font-size: .85rem; color: var(--muted); margin: 0 auto 1.5rem; padding: .4rem .8rem; border: 1px solid var(--border); border-radius: 999px; background: var(--card); width: max-content; }
      .theme-switch .track { width: 38px; height: 20px; border-radius: 999px; position: relative; background: linear-gradient(90deg, var(--accent), var(--accent-2)); }
      .theme-switch .track::after { content: ""; position: absolute; top: 2px; left: 2px; width: 16px; height: 16px; border-radius: 50%; background: #fff; transition: transform .3s ease; }
      #theme:checked ~ main .theme-switch .track::after { transform: translateX(18px); }
      .switch-row { display: flex; justify-content: center; }
    </style>
    </head>
    <body>
    <input type="checkbox" id="theme">
    <main>
      <div class="switch-row">
        <label class="theme-switch" for="theme">
          <span class="track"></span>Dark / Light
        </label>
      </div>
      <h2 id="choosing-a-vector-database">Choosing a vector database</h2>
      <p><strong>Short answer:</strong> start with pgvector if you already run Postgres; move to a dedicated store only when recall or latency forces you to.</p>
    
      <h3>Options compared</h3>
      <div class="t-scroll">
        <table class="t">
          <thead>
            <tr><th>Option</th><th>Best for</th><th>Watch out for</th></tr>
          </thead>
          <tbody>
            <tr><td>pgvector</td><td>teams already on Postgres</td><td>index build time above ~10M rows</td></tr>
            <tr><td>Qdrant</td><td>filtered search at scale</td><td>another service to operate</td></tr>
            <tr><td>Pinecone</td><td>zero ops</td><td>cost and vendor lock-in</td></tr>
          </tbody>
        </table>
      </div>
    
      <h3>Steps</h3>
      <ol>
        <li>Measure your corpus size and query rate.</li>
        <li>Prototype with pgvector and an HNSW index.</li>
        <li>Benchmark recall@10 against a held-out set.</li>
      </ol>
    
      <div class="callout callout-info"><span class="ic">&#8505;</span><div><strong>Note:</strong> embeddings from different models are not interchangeable - re-embed when you switch.</div></div>
    
      <p><code>CREATE INDEX ON items USING hnsw (embedding vector_cosine_ops);</code></p>
    </main>
    </body>
    </html>
    

    How the content was mapped

    • The level-2 heading became an anchored <h2>. The two level-3 headings became <h3>. There is only one level-2 heading, so the page has no table of contents.
    • The comparison table uses the template's striped table.t inside a horizontal-scroll wrapper.
    • The "Note:" blockquote became an info callout with a coloured left rail.
    • The SQL line was inline code in your source, so it stays inline <code> and is not a code block.
    • Your wording was not changed, and no badges, KPI cards or charts were added because the source has no status words or numeric series.

    response.html

    TSX · React component

    Generated

    Example file from a real run - the skill writes it into your workspace.

    $5/mo

    Subscription · always the latest version

    • Secure checkout via Stripe
    • Cancel anytime
    • Delivered live via MCP

    About this skill

    Takes a response (the last one, or any text you paste) and re-sets it as a single, self-contained HTML page that's easier to read and to share. Headings, lists, tables, code blocks, quotes and callouts become their semantic HTML, styled from one fixed stylesheet, so every page looks like it belongs to the same set.

    What it does

    You get one HTML file with no external assets and no script tags. It has a contents list when the answer has two or more sections, styled tables, highlighted callouts for notes and warnings, code blocks that keep their formatting, collapsible sections where they help, and a light/dark theme switch built in CSS only. The text is the original's: it reorganises how the answer looks, not what it says. In apps that render HTML inline it shows the page immediately; elsewhere it saves the file and gives you the source.

    Try it first

    After a long answer with a table and some steps, say "prettify this response as HTML". Open the file and flip the theme switch.

    What it needs

    Nothing beyond the response itself, plus write access if you want the file saved. No network access, no API keys and no build step.

    It formats; it doesn't fact-check or rewrite the answer. Charts are drawn only from numbers already in the text. The page is static: the theme switch, the collapsible sections and the contents work with CSS alone, so there's no search or copy button. It only runs when you ask for it, so it won't restyle every reply on its own.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 27 days ago

    Permissions

    Write Files
    Listed27 days ago

    What's inside

    Frequently Asked Questions

    Turns a long AI answer into one self-contained HTML page: headings, contents, styled tables, callouts and code, with a light/dark switch and no JavaScript

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use Llm Response Beautifier on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in Writing & Content