pagetrace 0.14.0 → 0.14.5

package/CHANGELOG.md

@@ -6,6 +6,85 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 While the version is below 1.0.0, breaking changes ship in a minor release.
 
+## [0.14.5] - 2026-09-08
+
+### Fixed
+
+- `extractLlmsTxt` used `Buffer.byteLength`, which does not exist in a browser,
+  so any page on a site that actually has an `/llms.txt` crashed the checker
+  with "Buffer is not defined". It counts UTF-8 bytes with `TextEncoder` now —
+  the same number, available everywhere. `extract.ts` is meant to be pure and
+  bundleable, and this was the one line that was not.
+- The scan log blamed the fetch for failures that happened after it, because
+  the error handler settled a fixed step rather than the one still running.
+
+## [0.14.4] - 2026-09-08
+
+### Added
+
+- Whole-site mode in the checker. It crawls from the sitemap the way
+  `pagetrace audit --url` does and runs every rule over the result, so the
+  cross-page findings a single page cannot produce — duplicate titles, hreflang
+  reciprocity, dead internal links, sitemap health — finally appear on the site.
+  Findings are grouped by code and message and list the routes they touch, which
+  is how the CLI's own report aggregates.
+
+### Fixed
+
+- Every request now has an eight-second deadline. A proxy that is down hangs
+  rather than refusing, and the chain sat waiting on it.
+- robots.txt and sitemap.xml were fetched through a reader that renders what it
+  fetches, so they came back as HTML documents and stopped looking like
+  themselves. Those requests now ask for text, and a sitemap whose XML did not
+  survive has its URLs recovered from the body rather than being given up on.
+- Switching scope left the other scope's score standing under a heading that no
+  longer described it.
+
+## [0.14.3] - 2026-09-08
+
+### Fixed
+
+- The checker could not fetch anything: it depended on one public CORS proxy,
+  and that proxy was returning Cloudflare 520s for every URL. It now tries
+  r.jina.ai, allorigins and codetabs in order and takes the first that answers
+  with HTML, so one service having a bad day no longer takes the feature down.
+
+### Added
+
+- A scan log. The check narrates itself as it runs — each request, its size and
+  which proxy served it, then the rule pass — instead of leaving a dead page
+  during the seconds it takes. A progress bar tracks the steps, the report dims
+  while it works, the score counts up to its value, and a category holding a
+  blocker gets a coloured edge.
+- Example URLs to try, so the page can be used without owning a site.
+
+## [0.14.2] - 2026-09-08
+
+### Added
+
+- An SEO checker page on the website: enter a URL and get all twenty-three
+  page-level checks scored and grouped, passes shown alongside failures. It
+  fetches robots.txt and llms.txt too, so the site-level checks answer from
+  evidence; anything not fetched reads "not checked" rather than passing or
+  failing.
+
+### Fixed
+
+- `npm run build:site` wrote into `docs/`, which tsup cleans, so building the
+  site deleted the site. Its output now has its own directory, `docs/assets/`,
+  which is the only thing the build can remove. The Pages deploy would have
+  published a lone JavaScript file.
+
+## [0.14.1] - 2026-09-08
+
+### Added
+
+- A website at [shyamexe.github.io/pagetrace](https://shyamexe.github.io/pagetrace/), built from
+  `docs/` and deployed by GitHub Pages. Its playground compiles `extract`, `audit` and `diff` into
+  the page via a new `src/browser.ts` entry, so it runs the same rules as the CLI rather than a
+  reimplementation of them. `npm run build:site` produces the bundle, and the Pages workflow rebuilds
+  it on every deploy so it cannot drift from the rules it claims to run.
+
 ## [0.14.0] - 2026-09-08
 
 ### Added
@@ -385,6 +464,11 @@ Initial release. `snapshot`, `check` and `audit` commands; filesystem and HTTP
 crawling; diff classified by transition; absolute, cross-page and hreflang audit
 rules; pretty, JSON, markdown, GitHub and HTML reporters.
 
+[0.14.5]: https://github.com/shyamexe/pagetrace/compare/v0.14.0...v0.14.5
+[0.14.4]: https://github.com/shyamexe/pagetrace/commit/ec2a5e2cd590
+[0.14.3]: https://github.com/shyamexe/pagetrace/commit/c45cfb49e265
+[0.14.2]: https://github.com/shyamexe/pagetrace/commit/b3aa77388ee9
+[0.14.1]: https://github.com/shyamexe/pagetrace/commit/7cc91383ed7a
 [0.14.0]: https://github.com/shyamexe/pagetrace/compare/v0.13.0...v0.14.0
 [0.13.0]: https://github.com/shyamexe/pagetrace/compare/v0.12.0...v0.13.0
 [0.12.0]: https://github.com/shyamexe/pagetrace/compare/v0.11.0...v0.12.0

package/README.md

@@ -19,6 +19,10 @@ npm install -D pagetrace
 
 Requires Node 20.19 or newer. No native modules, three small dependencies.
 
+## Playground
+
+Try the rules without installing anything: **[shyamexe.github.io/pagetrace](https://shyamexe.github.io/pagetrace/)**. The site compiles `extract`, `audit` and `diff` — the pure half of this package — into the page, so it runs the same rules the CLI does, in your browser, with nothing uploaded.
+
 ## Commands
 
 | Command | Answers | Crawls | Exit 1 when |

package/dist/cli.cjs

@@ -1549,7 +1549,7 @@ function extractLinks(html) {
 }
 function extractLlmsTxt(body) {
   const sections = body.split(/\r?\n/).filter((line) => line.startsWith("## ")).map((line) => line.slice(3).trim());
-  return { present: true, sections, bytes: Buffer.byteLength(body, "utf8") };
+  return { present: true, sections, bytes: new TextEncoder().encode(body).length };
 }
 var XML_ENTITIES = {
   amp: "&",
@@ -2210,11 +2210,11 @@ Commit both files, then run \`pagetrace check ${flags.dir ? `--dir ${flags.dir}`
 });
 cli.command("update", "Check npm for a newer pagetrace and install it").option("--check", "Only report whether an update exists").action(async (flags) => {
   const latest = await latestVersion("pagetrace");
-  if (!isNewer(latest, "0.14.0")) {
-    console.log(import_picocolors2.default.green(`pagetrace ${"0.14.0"} is the latest version.`));
+  if (!isNewer(latest, "0.14.5")) {
+    console.log(import_picocolors2.default.green(`pagetrace ${"0.14.5"} is the latest version.`));
     return;
   }
-  console.log(import_picocolors2.default.yellow(`Update available: ${"0.14.0"} \u2192 ${latest}`));
+  console.log(import_picocolors2.default.yellow(`Update available: ${"0.14.5"} \u2192 ${latest}`));
   if (flags.check) return;
   if (process.argv[1]?.startsWith(process.cwd())) {
     console.log(
@@ -2232,7 +2232,7 @@ cli.command("update", "Check npm for a newer pagetrace and install it").option("
   console.log(import_picocolors2.default.green(`Updated to pagetrace ${latest}.`));
 });
 cli.help();
-cli.version("0.14.0");
+cli.version("0.14.5");
 async function main() {
   try {
     cli.parse(process.argv, { run: false });

package/dist/cli.js

@@ -1526,7 +1526,7 @@ function extractLinks(html) {
 }
 function extractLlmsTxt(body) {
   const sections = body.split(/\r?\n/).filter((line) => line.startsWith("## ")).map((line) => line.slice(3).trim());
-  return { present: true, sections, bytes: Buffer.byteLength(body, "utf8") };
+  return { present: true, sections, bytes: new TextEncoder().encode(body).length };
 }
 var XML_ENTITIES = {
   amp: "&",
@@ -2187,11 +2187,11 @@ Commit both files, then run \`pagetrace check ${flags.dir ? `--dir ${flags.dir}`
 });
 cli.command("update", "Check npm for a newer pagetrace and install it").option("--check", "Only report whether an update exists").action(async (flags) => {
   const latest = await latestVersion("pagetrace");
-  if (!isNewer(latest, "0.14.0")) {
-    console.log(pc2.green(`pagetrace ${"0.14.0"} is the latest version.`));
+  if (!isNewer(latest, "0.14.5")) {
+    console.log(pc2.green(`pagetrace ${"0.14.5"} is the latest version.`));
     return;
   }
-  console.log(pc2.yellow(`Update available: ${"0.14.0"} \u2192 ${latest}`));
+  console.log(pc2.yellow(`Update available: ${"0.14.5"} \u2192 ${latest}`));
   if (flags.check) return;
   if (process.argv[1]?.startsWith(process.cwd())) {
     console.log(
@@ -2209,7 +2209,7 @@ cli.command("update", "Check npm for a newer pagetrace and install it").option("
   console.log(pc2.green(`Updated to pagetrace ${latest}.`));
 });
 cli.help();
-cli.version("0.14.0");
+cli.version("0.14.5");
 async function main() {
   try {
     cli.parse(process.argv, { run: false });

package/dist/index.cjs

@@ -988,7 +988,7 @@ function extractLinks(html) {
 }
 function extractLlmsTxt(body) {
   const sections = body.split(/\r?\n/).filter((line) => line.startsWith("## ")).map((line) => line.slice(3).trim());
-  return { present: true, sections, bytes: Buffer.byteLength(body, "utf8") };
+  return { present: true, sections, bytes: new TextEncoder().encode(body).length };
 }
 var XML_ENTITIES = {
   amp: "&",

package/dist/index.js

@@ -913,7 +913,7 @@ function extractLinks(html) {
 }
 function extractLlmsTxt(body) {
   const sections = body.split(/\r?\n/).filter((line) => line.startsWith("## ")).map((line) => line.slice(3).trim());
-  return { present: true, sections, bytes: Buffer.byteLength(body, "utf8") };
+  return { present: true, sections, bytes: new TextEncoder().encode(body).length };
 }
 var XML_ENTITIES = {
   amp: "&",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pagetrace",
-  "version": "0.14.0",
+  "version": "0.14.5",
   "description": "Baseline your site's SEO and AEO surface, diff every build against it, and fail CI on regressions.",
   "main": "./dist/index.cjs",
   "scripts": {
@@ -10,7 +10,8 @@
     "test:watch": "vitest",
     "coverage": "vitest run --coverage",
     "typecheck": "tsc --noEmit",
-    "prepublishOnly": "npm run typecheck && npm run test && npm run build"
+    "prepublishOnly": "npm run typecheck && npm run test && npm run build",
+    "build:site": "tsup src/browser.ts --format iife --global-name pagetrace --platform browser --minify --no-dts --out-dir docs/assets"
   },
   "keywords": [
     "seo",