diff --git a/AGENTS.md b/AGENTS.md
index d412f0d..9c72e3d 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -4,7 +4,7 @@
 
 This project provides small Firefox/Selenium browser utilities packaged by Nix:
 
-- `glimpse` - generic page utilities with subcommands, including provider-backed search
+- `glimpse` - headless browser CLI with subcommands for page extraction, JS execution, screenshots, and search
 
 Keep the tools simple, scriptable, and JSON-friendly.
 
@@ -18,7 +18,7 @@ npm run test:list
 node test/smoke.js <tag-or-name>
 ```
 
-For smoke testing without external network dependencies, use focused tags or scripts such as `npm run test:snapshot`, `npm run test:wait`, `npm run test:errors`, or `node test/smoke.js snapshot js`. Run `npm test` and `nix build .#default --no-link` when the change is broad, touches packaging, or needs full validation. Smoke tests require Firefox and geckodriver on `PATH` and use local `data:` HTML pages.
+For smoke testing without external network dependencies, use focused tags or scripts such as `npm run test:wait`, `npm run test:errors`, or `node test/smoke.js reader js`. Run `npm test` and `nix build .#default --no-link` when the change is broad, touches packaging, or needs full validation. Smoke tests require Firefox and geckodriver on `PATH` and use local `data:` HTML pages.
 
 Do not attempt a live Kagi test unless `KAGI_TOKEN` is available.
 
@@ -38,10 +38,9 @@ Do not attempt a live Kagi test unless `KAGI_TOKEN` is available.
 
 Current `glimpse` subcommands:
 
-- `snapshot <url>` - return an agent-friendly page snapshot as JSON
+- `reader <url>` - extract page content as Markdown (tries Firefox Reader View, falls back to raw Turndown conversion); supports `--no-reader` to skip Reader View, `--format=json` for structured output
 - `exec <url> --js=<code>` or `--script=<file>` - execute JavaScript and return the result
 - `screenshot <url> --output=<file>` - save a PNG screenshot
-- `reader <url>` - open Firefox Reader View and output readable content as Markdown
 - `search <query>` - search with a supported provider and output JSON results
 
 ## Runtime Requirements
diff --git a/src/index.ts b/src/index.ts
index 82d0061..c9a5916 100755
--- a/src/index.ts
+++ b/src/index.ts
@@ -21,6 +21,7 @@ interface ReaderArticle {
   sourceUrl?: string;
   finalUrl?: string;
   markdown?: string;
+  method?: "reader" | "raw";
 }
 
 // Parse CLI Args
@@ -87,10 +88,9 @@ function helpText() {
   return `Usage: glimpse <command> <url> [options]
 
 Commands:
-  snapshot <url> [options]    Return an agent-friendly page snapshot as JSON
+  reader <url> [options]      Extract page content as Markdown (Reader View with raw fallback)
   exec <url> [options]        Execute JavaScript on a page and return the result
   screenshot <url> [options]  Save a PNG screenshot of a page
-  reader <url> [options]      Extract Firefox Reader View content as Markdown
   search <query> [options]    Search using a supported provider and return JSON results
 
 Common Options:
@@ -114,6 +114,7 @@ Screenshot Options:
 Reader Options:
   --format=<format>           Output format: markdown, html, text, json (default: markdown)
   --output=<file>             Write output to a file
+  --no-reader                 Skip Reader View and use raw page extraction
 
 Search Options:
   --provider=<provider>       Search provider: kagi (default: config or kagi)
@@ -121,11 +122,12 @@ Search Options:
   --format=<format>           Output format: markdown, json (default: markdown)
 
 Examples:
-  glimpse snapshot https://example.com
+  glimpse reader https://example.com
+  glimpse reader https://example.com --no-reader
+  glimpse reader https://example.com/article --output=article.md
   glimpse exec https://example.com --js="return document.title"
   glimpse exec https://example.com --script=extract.js
   glimpse screenshot https://example.com --js="document.body.style.zoom = '80%'" --output=example.png
-  glimpse reader https://example.com/article --script=prepare.js --output=article.md
   KAGI_TOKEN=... glimpse search --provider=kagi "node.js browser automation"`;
 }
 
@@ -284,95 +286,6 @@ async function withPage(
   });
 }
 
-const snapshotScript = `
-const normalize = (value) => String(value || "").replace(/\\s+/g, " ").trim();
-const visibleText = (element) => normalize(element?.innerText || element?.textContent || "");
-const safeValue = (input) => ["password", "hidden"].includes(input.type) ? "" : input.value || "";
-const labelText = (input) => {
-  const labels = Array.from(input.labels || []).map((label) => visibleText(label)).filter(Boolean);
-  if (labels.length > 0) return labels.join(" ");
-
-  if (input.id) {
-    const label = Array.from(document.querySelectorAll("label[for]"))
-      .find((candidate) => candidate.getAttribute("for") === input.id);
-    if (label) return visibleText(label);
-  }
-
-  return "";
-};
-const inputSummary = (input) => ({
-  type: input.type || input.tagName.toLowerCase(),
-  name: input.name || "",
-  id: input.id || "",
-  placeholder: input.placeholder || "",
-  value: safeValue(input),
-  label: labelText(input),
-});
-const collectHeadings = () => {
-  try {
-    return Array.from(document.querySelectorAll("h1,h2,h3,h4,h5,h6,[role='heading']"))
-      .map((heading) => {
-        const tagLevel = heading.tagName.match(/^H([1-6])$/i)?.[1];
-        const ariaLevel = heading.getAttribute("aria-level");
-        const level = Number.parseInt(tagLevel || ariaLevel || "0", 10);
-        const text = visibleText(heading);
-
-        return text ? { level: level || null, text } : null;
-      })
-      .filter(Boolean);
-  } catch {
-    return [];
-  }
-};
-
-return {
-  text: normalize(document.body?.innerText || ""),
-  headings: collectHeadings(),
-  links: Array.from(document.querySelectorAll("a[href]"))
-    .map((link) => ({ text: visibleText(link), href: link.href }))
-    .filter((link) => link.text || link.href),
-  buttons: Array.from(document.querySelectorAll("button,input[type='button'],input[type='submit'],input[type='reset'],[role='button']"))
-    .map((button) => ({
-      text: visibleText(button) || button.value || button.getAttribute("aria-label") || "",
-      type: button.type || button.getAttribute("role") || "button",
-      name: button.name || "",
-      id: button.id || "",
-    }))
-    .filter((button) => button.text || button.name || button.id),
-  inputs: Array.from(document.querySelectorAll("input,textarea,select"))
-    .map(inputSummary),
-  forms: Array.from(document.querySelectorAll("form"))
-    .map((form) => ({
-      action: form.action || "",
-      method: (form.method || "get").toLowerCase(),
-      text: visibleText(form),
-      inputs: Array.from(form.querySelectorAll("input,textarea,select")).map(inputSummary),
-    })),
-};
-`;
-
-async function snapshotCommand() {
-  const [targetUrl] = getPositionalArgs();
-
-  if (!targetUrl) usage();
-
-  return withPage(targetUrl, async (driver: WebDriver) => {
-    // Capture Page Metadata
-    const [url, title, result] = await Promise.all([
-      driver.getCurrentUrl(),
-      driver.getTitle(),
-      driver.executeScript(snapshotScript),
-    ]);
-
-    return {
-      ok: true,
-      url,
-      title,
-      result,
-    };
-  });
-}
-
 async function execCommand() {
   const [targetUrl] = getPositionalArgs();
 
@@ -498,27 +411,19 @@ async function searchCommand() {
   }
 }
 
-async function readerCommand() {
-  const [targetUrl] = getPositionalArgs();
-  const outputPath = getOption("--output");
-  const format = getOption("--format") ?? "markdown";
+// Try Reader View Extraction
+async function tryReaderView(
+  driver: WebDriver,
+  finalUrl: string,
+  targetUrl: string,
+): Promise<ReaderArticle | null> {
+  const readerUrl = `about:reader?url=${encodeURIComponent(finalUrl)}`;
+  await driver.get(readerUrl);
 
-  if (!targetUrl) usage();
-
-  return withPage(targetUrl, async (driver: WebDriver) => {
-    // Capture Final Url
-    const finalUrl = await driver.getCurrentUrl();
-
-    // Open Firefox Reader View
-    const readerUrl = `about:reader?url=${encodeURIComponent(finalUrl)}`;
-    await driver.get(readerUrl);
-
-    // Wait For Reader Content
-    let article: ReaderArticle;
-    try {
-      article = await driver.wait(
-        async () => {
-          return driver.executeScript(`
+  try {
+    return await driver.wait(
+      async () => {
+        return driver.executeScript(`
           const content = document.querySelector("#moz-reader-content, .moz-reader-content");
           const error = document.querySelector(".reader-error");
           const text = content?.innerText?.trim() || "";
@@ -540,15 +445,67 @@ async function readerCommand() {
 
           return null;
         `);
-        },
-        timeoutMs,
-        `No readable article content found for URL: ${targetUrl}`,
-      );
-    } catch (err) {
-      cliError("TIMEOUT", err.message);
+      },
+      timeoutMs,
+      `No readable article content found for URL: ${targetUrl}`,
+    );
+  } catch {
+    return null;
+  }
+}
+
+// Raw Page Extraction Fallback
+async function extractRawPage(
+  driver: WebDriver,
+  originalUrl?: string,
+): Promise<ReaderArticle> {
+  // Navigate Back If Needed (e.g. after failed Reader View)
+  if (originalUrl) {
+    await driver.get(originalUrl);
+  }
+
+  const result = (await driver.executeScript(`
+    return {
+      title: document.title || "",
+      html: document.body?.innerHTML || "",
+      text: document.body?.innerText?.trim() || "",
+    };
+  `)) as { title: string; html: string; text: string };
+
+  return {
+    title: result.title,
+    html: result.html,
+    text: result.text,
+  };
+}
+
+async function readerCommand() {
+  const [targetUrl] = getPositionalArgs();
+  const outputPath = getOption("--output");
+  const format = getOption("--format") ?? "markdown";
+  const skipReader = args.includes("--no-reader");
+
+  if (!targetUrl) usage();
+
+  return withPage(targetUrl, async (driver: WebDriver) => {
+    const finalUrl = await driver.getCurrentUrl();
+
+    // Extract Page Content
+    let article: ReaderArticle;
+    if (!skipReader) {
+      article = await tryReaderView(driver, finalUrl, targetUrl);
     }
 
-    // Render Output
+    // Fallback To Raw Extraction
+    if (!article) {
+      // Navigate back only if Reader View was attempted
+      article = await extractRawPage(driver, skipReader ? undefined : finalUrl);
+      article.method = "raw";
+    } else {
+      article.method = "reader";
+    }
+
+    // Build Output
     article.sourceUrl = targetUrl;
     article.finalUrl = finalUrl;
     article.markdown = articleToMarkdown(article);
@@ -583,8 +540,6 @@ async function main() {
   appConfig = loadConfig({ path: configPath });
 
   switch (command) {
-    case "snapshot":
-      return snapshotCommand();
     case "exec":
       return execCommand();
     case "screenshot":
diff --git a/test/smoke.js b/test/smoke.js
index 1367215..5944a61 100755
--- a/test/smoke.js
+++ b/test/smoke.js
@@ -1,4 +1,5 @@
 #!/usr/bin/env node
+// @ts-nocheck
 
 import {
   mkdtempSync,
@@ -73,7 +74,7 @@ test("no args prints help", ["help", "cli"], () => {
 
   assert.equal(result.status, 0, result.stderr || result.stdout);
   assert.match(result.stdout, /Usage: glimpse <command> <url> \[options\]/);
-  assert.match(result.stdout, /snapshot <url>/);
+  assert.match(result.stdout, /reader <url>/);
   assert.equal(result.stderr, "");
 });
 
@@ -86,48 +87,50 @@ test("help flag prints help", ["help", "cli"], () => {
   assert.equal(result.stderr, "");
 });
 
-test("snapshot returns page metadata and content", ["snapshot"], () => {
-  const output = expectSuccess([
-    "snapshot",
+test("reader extracts page content as markdown", ["reader"], () => {
+  const result = runCli([
+    "reader",
     dataHtml(
-      '<title>Hello</title><h1>Main</h1><a href="/x">X</a><button>Go</button>',
+      '<title>Hello</title><h1>Main</h1><p>Some text</p><a href="https://example.com">Link</a>',
     ),
+    "--no-reader",
   ]);
 
-  assert.equal(output.ok, true);
-  assert.equal(output.title, "Hello");
-  assert.equal(typeof output.elapsedMs, "number");
-  assert.deepEqual(output.result.headings, [{ level: 1, text: "Main" }]);
-  assert.deepEqual(output.result.links, [{ href: "/x", text: "X" }]);
-  assert.equal(output.result.buttons[0].text, "Go");
-  assert.match(output.result.text, /Main/);
+  assert.equal(result.status, 0, result.stderr || result.stdout);
+  const output = result.stdout.trim();
+  assert.match(output, /# Main/);
+  assert.match(output, /Some text/);
+  assert.match(output, /\[Link\]\(https:\/\/example\.com\/??\)/);
 });
 
-test("snapshot extracts aria headings", ["snapshot"], () => {
+test("reader returns json format with method field", ["reader"], () => {
   const output = expectSuccess([
-    "snapshot",
-    dataHtml(
-      '<title>Hello</title><div role="heading" aria-level="2">ARIA</div>',
-    ),
+    "reader",
+    dataHtml("<title>Hello</title><h1>Main</h1><p>World</p>"),
+    "--no-reader",
+    "--format=json",
   ]);
 
-  assert.equal(output.ok, true);
-  assert.deepEqual(output.result.headings, [{ level: 2, text: "ARIA" }]);
+  assert.equal(output.title, "Hello");
+  assert.equal(output.method, "raw");
+  assert.equal(typeof output.markdown, "string");
+  assert.match(output.markdown, /# Main/);
+  assert.match(output.text, /Main/);
 });
 
 test(
-  "snapshot runs top-level javascript before extraction",
-  ["snapshot", "js"],
+  "reader runs top-level javascript before extraction",
+  ["reader", "js"],
   () => {
-    const output = expectSuccess([
-      "snapshot",
+    const result = runCli([
+      "reader",
       dataHtml("<title>Hello</title><h1>Old</h1>"),
+      "--no-reader",
       "--js=document.querySelector('h1').textContent = 'New'",
     ]);
 
-    assert.equal(output.ok, true);
-    assert.deepEqual(output.result.headings, [{ level: 1, text: "New" }]);
-    assert.equal(output.result.text, "New");
+    assert.equal(result.status, 0, result.stderr || result.stdout);
+    assert.match(result.stdout, /# New/);
   },
 );
 
@@ -182,8 +185,9 @@ test(
     writeFileSync(configPath, "not json");
 
     const output = expectFailure([
-      "snapshot",
+      "reader",
       dataHtml("<title>Hello</title>"),
+      "--no-reader",
       `--config=${configPath}`,
     ]);
 
@@ -201,8 +205,9 @@ test(
     writeFileSync(configPath, JSON.stringify({ search: { provider: 42 } }));
 
     const output = expectFailure([
-      "snapshot",
+      "reader",
       dataHtml("<title>Hello</title>"),
+      "--no-reader",
       `--config=${configPath}`,
     ]);
 
@@ -218,17 +223,17 @@ test("empty home config is accepted", ["config"], () => {
   mkdirSync(configDir, { recursive: true });
   writeFileSync(join(configDir, "config.json"), "{}");
 
-  const output = expectSuccess(
-    ["snapshot", dataHtml("<title>Hello</title><h1>Main</h1>")],
+  const result = runCli(
+    ["reader", dataHtml("<title>Hello</title><h1>Main</h1>"), "--no-reader"],
     { env: { ...process.env, XDG_CONFIG_HOME: configHome } },
   );
 
-  assert.equal(output.ok, true);
-  assert.equal(output.title, "Hello");
+  assert.equal(result.status, 0, result.stderr || result.stdout);
+  assert.match(result.stdout, /# Main/);
 });
 
 test("unknown command returns structured error", ["errors", "cli"], () => {
-  const output = expectFailure(["nope", dataHtml("<title>Hello</title>")]);
+  const output = expectFailure(["nope", dataHtml("<title>Hello</title>"), "--no-reader"]);
 
   assert.equal(output.ok, false);
   assert.equal(output.error.code, "UNKNOWN_COMMAND");
@@ -241,8 +246,9 @@ test(
   ["errors", "timeout"],
   () => {
     const output = expectFailure([
-      "snapshot",
+      "reader",
       dataHtml("<title>Hello</title>"),
+      "--no-reader",
       "--timeout=abc",
     ]);
 
@@ -255,8 +261,9 @@ test(
 
 test("invalid wait-until returns invalid option", ["errors", "wait"], () => {
   const output = expectFailure([
-    "snapshot",
+    "reader",
     dataHtml("<title>Hello</title>"),
+    "--no-reader",
     "--wait-until=loaded",
   ]);
 
@@ -266,20 +273,22 @@ test("invalid wait-until returns invalid option", ["errors", "wait"], () => {
 });
 
 test("wait-js succeeds when condition is true", ["wait"], () => {
-  const output = expectSuccess([
-    "snapshot",
-    dataHtml("<title>Hello</title>"),
+  const result = runCli([
+    "reader",
+    dataHtml("<title>Hello</title><h1>Main</h1>"),
+    "--no-reader",
     '--wait-js=return document.title === "Hello"',
   ]);
 
-  assert.equal(output.ok, true);
-  assert.equal(output.title, "Hello");
+  assert.equal(result.status, 0, result.stderr || result.stdout);
+  assert.match(result.stdout, /# Main/);
 });
 
 test("wait-js timeout returns wait timeout", ["wait", "errors"], () => {
   const output = expectFailure([
-    "snapshot",
+    "reader",
     dataHtml("<title>Hello</title>"),
+    "--no-reader",
     "--wait-js=return false",
     "--timeout=1",
   ]);
@@ -296,8 +305,9 @@ test(
   ["wait", "errors", "js"],
   () => {
     const output = expectFailure([
-      "snapshot",
+      "reader",
       dataHtml("<title>Hello</title>"),
+      "--no-reader",
       '--wait-js=throw new Error("boom")',
     ]);
 
@@ -313,8 +323,9 @@ test(
   ["errors", "js"],
   () => {
     const output = expectFailure([
-      "snapshot",
+      "reader",
       dataHtml("<title>Hello</title>"),
+      "--no-reader",
       '--js=throw new Error("boom")',
     ]);
 
diff --git a/tsconfig.json b/tsconfig.json
index 5ee02d9..6fc29f4 100644
--- a/tsconfig.json
+++ b/tsconfig.json
@@ -11,5 +11,6 @@
     "forceConsistentCasingInFileNames": true,
     "skipLibCheck": true
   },
-  "include": ["src/**/*.ts"]
+  "include": ["src/**/*.ts"],
+  "exclude": ["test", "node_modules", "dist"]
 }