diff --git a/package-lock.json b/package-lock.json
index 2bdc8d7..f30ebca 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -8,6 +8,7 @@
 			"name": "squalk",
 			"version": "0.0.1",
 			"dependencies": {
+				"@djot/djot": "^0.3.2",
 				"@nostr/gadgets": "npm:@jsr/nostr__gadgets@^0.1.8",
 				"@nostr/tools": "npm:@jsr/nostr__tools@^2.23.3"
 			},
@@ -37,6 +38,18 @@
 				"vite": "^8.0.7"
 			}
 		},
+		"node_modules/@djot/djot": {
+			"version": "0.3.2",
+			"resolved": "https://registry.npmjs.org/@djot/djot/-/djot-0.3.2.tgz",
+			"integrity": "sha512-joMKR24B8rxueyFiJbpZAqEiypjvOyzTxzkhyr0q5mM/sUBaOD3unna/9IxtOotFugViyYlkIRaiXg3xM//zxg==",
+			"license": "MIT",
+			"bin": {
+				"djot": "lib/cli.js"
+			},
+			"engines": {
+				"node": ">=17.0.0"
+			}
+		},
 		"node_modules/@emnapi/core": {
 			"version": "1.10.0",
 			"resolved": "https://registry.npmjs.org/@emnapi/core/-/core-1.10.0.tgz",
diff --git a/package.json b/package.json
index 0cfce2a..7ae9eea 100644
--- a/package.json
+++ b/package.json
@@ -39,6 +39,7 @@
 		"vite": "^8.0.7"
 	},
 	"dependencies": {
+		"@djot/djot": "^0.3.2",
 		"@nostr/gadgets": "npm:@jsr/nostr__gadgets@^0.1.8",
 		"@nostr/tools": "npm:@jsr/nostr__tools@^2.23.3"
 	}
diff --git a/src/lib/components/PostContent.svelte b/src/lib/components/PostContent.svelte
index 4b6e6e8..960a451 100644
--- a/src/lib/components/PostContent.svelte
+++ b/src/lib/components/PostContent.svelte
@@ -1,6 +1,15 @@
 <script lang="ts">
   import { loadNostrUser, type NostrUser } from "@nostr/gadgets/metadata";
   import * as nip19 from "@nostr/tools/nip19";
+  import { parse } from "@djot/djot";
+  import type {
+    Block as DjBlock,
+    Inline as DjInline,
+    Link as DjLink,
+    Image as DjImage,
+    Table as DjTable,
+    Reference,
+  } from "@djot/djot";
 
   type Props = {
     content: string;
@@ -130,14 +139,34 @@
     | { type: "link"; href: string; label: string }
     | { type: "mention"; pubkey: string; entity: string; fallback: string }
     | { type: "entity"; entity: string; label: string }
-    | { type: "thread-quote"; pubkey: string; eventId: string };
+    | { type: "thread-quote"; pubkey: string; eventId: string }
+    | { type: "strong"; children: Inline[] }
+    | { type: "em"; children: Inline[] }
+    | { type: "del"; children: Inline[] }
+    | { type: "code"; value: string }
+    | { type: "br" }
+    | { type: "image"; src: string; alt: string };
+
+  type Align = "default" | "left" | "right" | "center";
+  type TableCell = { align: Align; inlines: Inline[] };
+  type TableRow = { head: boolean; cells: TableCell[] };
+  type ListItem = { checked: boolean | null; blocks: Block[] };
 
   type Block =
-    | { type: "image"; value: string }
+    | { type: "image"; src: string; alt: string }
     | { type: "para"; inlines: Inline[] }
-    | { type: "blockquote"; blocks: Block[] };
-
-  const BLOCKQUOTE_LINE_RE = /^>\s?(.*)$/;
+    | { type: "blockquote"; blocks: Block[] }
+    | { type: "heading"; level: number; inlines: Inline[] }
+    | { type: "code"; lang: string; value: string }
+    | { type: "hr" }
+    | {
+        type: "list";
+        ordered: boolean;
+        start: number;
+        tight: boolean;
+        items: ListItem[];
+      }
+    | { type: "table"; rows: TableRow[] };
 
   function decodeNostrInline(entity: string): Inline | null {
     try {
@@ -181,8 +210,9 @@
     return null;
   }
 
-  // Inline-only tokenizer: produces inline tokens for a single line of text.
-  // Used inside blockquote lines, where images are not supported.
+  // Leaf text pass: scans a plain-text run (a Djot `str` node) for nostr
+  // entities and bare URLs. Never runs on code — verbatim/code_block stay
+  // literal.
   function tokenizeInline(text: string): Inline[] {
     const inlines: Inline[] = [];
     let last = 0;
@@ -217,108 +247,293 @@
     return inlines;
   }
 
-  function isBlockquoteParagraph(text: string): boolean {
-    const lines = text.split("\n");
-    let hasQuoteLine = false;
-    for (const l of lines) {
-      if (l.length === 0) continue;
-      if (!BLOCKQUOTE_LINE_RE.test(l)) return false;
-      hasQuoteLine = true;
-    }
-    return hasQuoteLine;
+  // Whitelist link schemes: user-controlled hrefs are the one place a raw
+  // value reaches an <a>. Anything not http(s)/mailto renders as plain text.
+  function safeHref(dest: string): string | null {
+    const d = dest.trim();
+    if (/^https?:\/\//i.test(d)) return d;
+    if (/^mailto:/i.test(d)) return d;
+    if (/^[a-z][a-z0-9+.-]*:/i.test(d)) return null;
+    if (/\S\.\S/.test(d)) return `https://${d}`;
+    return null;
   }
 
-  function tokenizeBlockquote(text: string): Block {
-    // Strip the > prefix per line, then route through the same paragraph
-    // pipeline so blockquotes get image support, pre-wrap newlines, and
-    // consistent paragraph handling.
-    const inner = text
-      .split("\n")
-      .map((l) => {
-        const m = l.match(BLOCKQUOTE_LINE_RE);
-        return m ? m[1] : l;
-      })
-      .join("\n");
-    const blocks: Block[] = [];
-    for (const para of inner.split(/\n{2,}/)) {
-      for (const b of tokenize(para)) blocks.push(b);
-    }
-    return { type: "blockquote", blocks };
+  function safeImg(dest: string): string | null {
+    const d = dest.trim();
+    if (/^https?:\/\//i.test(d)) return d;
+    if (/^[a-z][a-z0-9+.-]*:/i.test(d)) return null;
+    if (/\S\.\S/.test(d)) return `https://${d}`;
+    return null;
   }
 
-  function tokenize(text: string): Block[] {
-    if (isBlockquoteParagraph(text)) return [tokenizeBlockquote(text)];
+  function inlineText(nodes: DjInline[]): string {
+    let s = "";
+    for (const n of nodes) {
+      if ("text" in n && typeof n.text === "string") s += n.text;
+      else if ("children" in n) s += inlineText(n.children as DjInline[]);
+    }
+    return s;
+  }
+
+  // Reference-style link/image targets, populated per parse.
+  let references: Record<string, Reference> = {};
+
+  function resolveDest(node: DjLink | DjImage): string | undefined {
+    if (node.destination !== undefined) return node.destination;
+    if (node.reference !== undefined)
+      return references[node.reference]?.destination;
+    return undefined;
+  }
+
+  function convertLink(node: DjLink): Inline {
+    const label = inlineText(node.children) || node.destination || "";
+    const dest = resolveDest(node);
+    if (!dest) return { type: "text", value: label };
+    if (/^nostr:/i.test(dest)) {
+      const entity = dest.slice(6).toLowerCase();
+      try {
+        nip19.decode(entity);
+        return { type: "link", href: `https://njump.me/${entity}`, label };
+      } catch {
+        return { type: "text", value: label };
+      }
+    }
+    const href = safeHref(dest);
+    if (!href) return { type: "text", value: label };
+    return { type: "link", href, label };
+  }
+
+  function convertImage(node: DjImage): Inline {
+    const alt = inlineText(node.children);
+    const dest = resolveDest(node);
+    const src = dest ? safeImg(dest) : null;
+    if (!src) return { type: "text", value: alt };
+    return { type: "image", src, alt };
+  }
+
+  function convertInline(node: DjInline): Inline[] {
+    switch (node.tag) {
+      case "str":
+        return tokenizeInline(node.text);
+      case "soft_break":
+        return [{ type: "text", value: " " }];
+      case "hard_break":
+        return [{ type: "br" }];
+      case "non_breaking_space":
+        return [{ type: "text", value: " " }];
+      case "verbatim":
+        return [{ type: "code", value: node.text }];
+      case "strong":
+        return [{ type: "strong", children: convertInlines(node.children) }];
+      case "emph":
+        return [{ type: "em", children: convertInlines(node.children) }];
+      case "delete":
+        return [{ type: "del", children: convertInlines(node.children) }];
+      case "mark":
+      case "insert":
+      case "span":
+      case "superscript":
+      case "subscript":
+      case "double_quoted":
+      case "single_quoted":
+        return convertInlines(node.children);
+      case "link":
+        return [convertLink(node)];
+      case "image":
+        return [convertImage(node)];
+      case "url": {
+        const href = safeHref(node.text);
+        return href
+          ? [{ type: "link", href, label: node.text }]
+          : [{ type: "text", value: node.text }];
+      }
+      case "email":
+        return [
+          { type: "link", href: `mailto:${node.text}`, label: node.text },
+        ];
+      case "smart_punctuation":
+        return [{ type: "text", value: node.text }];
+      case "symb":
+        return [{ type: "text", value: `:${node.alias}:` }];
+      case "raw_inline":
+      case "inline_math":
+      case "display_math":
+        return [{ type: "text", value: node.text }];
+      default:
+        return [];
+    }
+  }
+
+  function convertInlines(nodes: DjInline[]): Inline[] {
+    const out: Inline[] = [];
+    for (const n of nodes) out.push(...convertInline(n));
+    return out;
+  }
+
+  // Split an inline run so images (markdown images or bare image URLs)
+  // become their own block, matching the standalone-image convention.
+  function splitImages(inlines: Inline[]): Block[] {
     const blocks: Block[] = [];
-    let inlines: Inline[] = [];
+    let cur: Inline[] = [];
     const flush = () => {
-      // Trim leading/trailing newlines so block images don't carry an extra
-      // visible line break under whitespace-pre-wrap.
-      while (inlines.length > 0) {
-        const first = inlines[0];
-        if (first.type !== "text") break;
-        const trimmed = first.value.replace(/^\n+/, "");
-        if (trimmed === "") inlines.shift();
+      while (cur.length) {
+        const f = cur[0];
+        if (f.type !== "text") break;
+        const t = f.value.replace(/^\s+/, "");
+        if (t === "") cur.shift();
         else {
-          inlines[0] = { type: "text", value: trimmed };
+          cur[0] = { type: "text", value: t };
           break;
         }
       }
-      while (inlines.length > 0) {
-        const last = inlines[inlines.length - 1];
-        if (last.type !== "text") break;
-        const trimmed = last.value.replace(/\n+$/, "");
-        if (trimmed === "") inlines.pop();
+      while (cur.length) {
+        const l = cur[cur.length - 1];
+        if (l.type !== "text") break;
+        const t = l.value.replace(/\s+$/, "");
+        if (t === "") cur.pop();
         else {
-          inlines[inlines.length - 1] = { type: "text", value: trimmed };
+          cur[cur.length - 1] = { type: "text", value: t };
           break;
         }
       }
-      if (inlines.some((i) => i.type !== "text" || i.value.trim()))
-        blocks.push({ type: "para", inlines });
-      inlines = [];
+      if (cur.some((i) => i.type !== "text" || i.value.trim()))
+        blocks.push({ type: "para", inlines: cur });
+      cur = [];
     };
-    let last = 0;
-    for (const m of text.matchAll(URL_RE)) {
-      const start = m.index ?? 0;
-      let url = m[0];
-      const isNostr = /^nostr:/i.test(url);
-      let trailing = "";
-      if (!isNostr) {
-        const trail = url.match(TRAILING_PUNCT_RE);
-        if (trail) {
-          trailing = trail[0];
-          url = url.slice(0, -trailing.length);
-        }
-      }
-      if (start > last)
-        inlines.push({ type: "text", value: text.slice(last, start) });
-      if (isNostr) {
-        const entity = url.slice(6).toLowerCase();
-        const decoded = decodeNostrInline(entity);
-        if (decoded) inlines.push(decoded);
-        else inlines.push({ type: "text", value: m[0] });
-      } else {
-        const href = /^https?:\/\//i.test(url) ? url : `https://${url}`;
-        if (IMG_EXT_RE.test(url)) {
-          flush();
-          blocks.push({ type: "image", value: href });
-          if (trailing) inlines.push({ type: "text", value: trailing });
-        } else {
-          inlines.push({ type: "link", href, label: url });
-          if (trailing) inlines.push({ type: "text", value: trailing });
-        }
-      }
-      last = start + m[0].length;
+    for (const inl of inlines) {
+      if (inl.type === "image") {
+        flush();
+        blocks.push({ type: "image", src: inl.src, alt: inl.alt });
+      } else if (inl.type === "link" && IMG_EXT_RE.test(inl.href)) {
+        flush();
+        blocks.push({ type: "image", src: inl.href, alt: "" });
+      } else cur.push(inl);
     }
-    if (last < text.length)
-      inlines.push({ type: "text", value: text.slice(last) });
     flush();
     return blocks;
   }
 
-  const paragraphs = $derived(
-    content.split(/\n{2,}/).map((para) => tokenize(para)),
-  );
+  function convertTable(node: DjTable): Block {
+    const rows: TableRow[] = [];
+    for (const child of node.children) {
+      if (child.tag !== "row") continue;
+      rows.push({
+        head: child.head,
+        cells: child.children.map((cell) => ({
+          align: cell.align,
+          inlines: convertInlines(cell.children),
+        })),
+      });
+    }
+    return { type: "table", rows };
+  }
+
+  function convertBlock(node: DjBlock): Block[] {
+    switch (node.tag) {
+      case "para":
+        return splitImages(convertInlines(node.children));
+      case "heading":
+        return [
+          {
+            type: "heading",
+            level: node.level,
+            inlines: convertInlines(node.children),
+          },
+        ];
+      case "thematic_break":
+        return [{ type: "hr" }];
+      case "section":
+      case "div":
+        return convertBlocks(node.children);
+      case "code_block":
+        return [
+          {
+            type: "code",
+            lang: node.lang ?? "",
+            value: node.text.replace(/\n$/, ""),
+          },
+        ];
+      case "raw_block":
+        return [
+          { type: "para", inlines: [{ type: "text", value: node.text }] },
+        ];
+      case "block_quote":
+        return [{ type: "blockquote", blocks: convertBlocks(node.children) }];
+      case "bullet_list":
+        return [
+          {
+            type: "list",
+            ordered: false,
+            start: 1,
+            tight: node.tight,
+            items: node.children.map((li) => ({
+              checked: null,
+              blocks: convertBlocks(li.children),
+            })),
+          },
+        ];
+      case "ordered_list":
+        return [
+          {
+            type: "list",
+            ordered: true,
+            start: node.start ?? 1,
+            tight: node.tight,
+            items: node.children.map((li) => ({
+              checked: null,
+              blocks: convertBlocks(li.children),
+            })),
+          },
+        ];
+      case "task_list":
+        return [
+          {
+            type: "list",
+            ordered: false,
+            start: 1,
+            tight: node.tight,
+            items: node.children.map((li) => ({
+              checked: li.checkbox === "checked",
+              blocks: convertBlocks(li.children),
+            })),
+          },
+        ];
+      case "table":
+        return [convertTable(node)];
+      case "definition_list": {
+        const out: Block[] = [];
+        for (const item of node.children) {
+          const [term, def] = item.children;
+          out.push({
+            type: "para",
+            inlines: [
+              { type: "strong", children: convertInlines(term.children) },
+            ],
+          });
+          out.push(...convertBlocks(def.children));
+        }
+        return out;
+      }
+      default:
+        return [];
+    }
+  }
+
+  function convertBlocks(nodes: DjBlock[]): Block[] {
+    const out: Block[] = [];
+    for (const n of nodes) out.push(...convertBlock(n));
+    return out;
+  }
+
+  const blocks = $derived.by<Block[]>(() => {
+    try {
+      const doc = parse(content);
+      references = { ...doc.references, ...doc.autoReferences };
+      return convertBlocks(doc.children);
+    } catch {
+      return [{ type: "para", inlines: [{ type: "text", value: content }] }];
+    }
+  });
 
   let resolvedUsers = $state<Record<string, NostrUser>>({});
 
@@ -328,15 +543,27 @@
       for (const inline of inlines) {
         if (inline.type === "mention" || inline.type === "thread-quote")
           seen.add(inline.pubkey);
+        else if (
+          inline.type === "strong" ||
+          inline.type === "em" ||
+          inline.type === "del"
+        )
+          collect(inline.children);
       }
     };
     const visit = (blocks: Block[]) => {
       for (const block of blocks) {
-        if (block.type === "para") collect(block.inlines);
+        if (block.type === "para" || block.type === "heading")
+          collect(block.inlines);
         else if (block.type === "blockquote") visit(block.blocks);
+        else if (block.type === "list")
+          for (const item of block.items) visit(item.blocks);
+        else if (block.type === "table")
+          for (const row of block.rows)
+            for (const cell of row.cells) collect(cell.inlines);
       }
     };
-    for (const blocks of paragraphs) visit(blocks);
+    visit(blocks);
     for (const pubkey of seen) {
       if (resolvedUsers[pubkey] || profiles[pubkey]) continue;
       loadNostrUser(pubkey).then((u) => {
@@ -347,7 +574,7 @@
 </script>
 
 {#snippet renderInlines(inlines: Inline[])}
-  {#each inlines as inline}
+  {#each inlines as inline (inline)}
     {#if inline.type === "link"}
       <a
         href={inline.href}
@@ -390,16 +617,54 @@
         rel="noopener noreferrer"
         class="text-brand hover:underline break-all">{inline.label}</a
       >
+    {:else if inline.type === "strong"}
+      <strong>{@render renderInlines(inline.children)}</strong>
+    {:else if inline.type === "em"}
+      <em>{@render renderInlines(inline.children)}</em>
+    {:else if inline.type === "del"}
+      <del>{@render renderInlines(inline.children)}</del>
+    {:else if inline.type === "code"}
+      <code>{inline.value}</code>
+    {:else if inline.type === "br"}
+      <br />
+    {:else if inline.type === "image"}
+      <img
+        src={inline.src}
+        alt={inline.alt}
+        loading="lazy"
+        class="inline-block max-h-[1.5em] align-text-bottom"
+      />
     {:else}{inline.value}{/if}
   {/each}
 {/snippet}
 
+{#snippet renderListItem(item: ListItem, tight: boolean)}
+  <li>
+    {#if item.checked !== null}
+      <input
+        type="checkbox"
+        checked={item.checked}
+        disabled
+        aria-label={item.checked ? "Completed" : "Not completed"}
+        class="mr-1 align-middle"
+      />
+    {/if}
+    {#each item.blocks as b (b)}
+      {#if tight && b.type === "para"}
+        {@render renderInlines(b.inlines)}
+      {:else}
+        {@render renderBlocks([b])}
+      {/if}
+    {/each}
+  </li>
+{/snippet}
+
 {#snippet renderBlocks(blocks: Block[])}
-  {#each blocks as block}
+  {#each blocks as block (block)}
     {#if block.type === "image"}
       <img
-        src={block.value}
-        alt=""
+        src={block.src}
+        alt={block.alt}
         loading="lazy"
         class="block mx-auto w-full max-h-[80vh] object-contain rounded"
       />
@@ -409,16 +674,72 @@
       >
         {@render renderBlocks(block.blocks)}
       </blockquote>
+    {:else if block.type === "heading"}
+      {@const tag = `h${Math.min(block.level + 2, 6)}`}
+      <svelte:element this={tag}>
+        {@render renderInlines(block.inlines)}
+      </svelte:element>
+    {:else if block.type === "code"}
+      <pre class="overflow-x-auto"><code>{block.value}</code></pre>
+    {:else if block.type === "hr"}
+      <hr />
+    {:else if block.type === "list"}
+      {#if block.ordered}
+        <ol start={block.start !== 1 ? block.start : undefined}>
+          {#each block.items as item (item)}
+            {@render renderListItem(item, block.tight)}
+          {/each}
+        </ol>
+      {:else}
+        <ul>
+          {#each block.items as item (item)}
+            {@render renderListItem(item, block.tight)}
+          {/each}
+        </ul>
+      {/if}
+    {:else if block.type === "table"}
+      {@const headRows = block.rows.filter((r) => r.head)}
+      {@const bodyRows = block.rows.filter((r) => !r.head)}
+      <div class="overflow-x-auto my-4">
+        <table>
+          {#if headRows.length}
+            <thead>
+              {#each headRows as row (row)}
+                <tr>
+                  {#each row.cells as cell (cell)}
+                    <th
+                      style:text-align={cell.align !== "default"
+                        ? cell.align
+                        : null}>{@render renderInlines(cell.inlines)}</th
+                    >
+                  {/each}
+                </tr>
+              {/each}
+            </thead>
+          {/if}
+          <tbody>
+            {#each bodyRows as row (row)}
+              <tr>
+                {#each row.cells as cell (cell)}
+                  <td
+                    style:text-align={cell.align !== "default"
+                      ? cell.align
+                      : null}>{@render renderInlines(cell.inlines)}</td
+                  >
+                {/each}
+              </tr>
+            {/each}
+          </tbody>
+        </table>
+      </div>
     {:else}
-      <p class="whitespace-pre-wrap">{@render renderInlines(block.inlines)}</p>
+      <p>{@render renderInlines(block.inlines)}</p>
     {/if}
   {/each}
 {/snippet}
 
 <div
-  class="prose leading-5 max-w-none text-neutral-700 [&_p]:my-3 [&_img]:my-5 [&_blockquote_p]:before:content-none [&_blockquote_p]:after:content-none"
+  class="prose leading-5 max-w-none text-neutral-700 [&_p]:my-3 [&_img]:my-5 [&_table]:my-0 [&_pre]:bg-neutral-100 [&_pre]:text-neutral-800 [&_:not(pre)>code]:rounded [&_:not(pre)>code]:bg-neutral-100 [&_:not(pre)>code]:px-1 [&_:not(pre)>code]:py-0.5 [&_blockquote_p]:before:content-none [&_blockquote_p]:after:content-none [&_code]:before:content-none [&_code]:after:content-none"
 >
-  {#each paragraphs as blocks}
-    {@render renderBlocks(blocks)}
-  {/each}
+  {@render renderBlocks(blocks)}
 </div>