fix(seo): ship a robots.txt so the production smoke test passes

public/robots.txtadded+24-0

Changes to public/robots.txt

		@@ -0,0 +1,24 @@
	1	+# https://commshq.org — crawl policy.
	2	+#
	3	+# The auth-walled routes below also send `robots: noindex, nofollow` from their
	4	+# own useSeoMeta call, which is what keeps them out of an index once fetched.
	5	+# This file stops well-behaved crawlers spending the request at all, and is the
	6	+# only place that can speak for routes that never render HTML.
	7	+#
	8	+# No sitemap is published yet (sitemap.xml is a 404). Add a Sitemap: line here
	9	+# when one exists rather than pointing crawlers at a missing file.
	10	+
	11	+User-agent: *
	12	+
	13	+Disallow: /dashboard
	14	+Disallow: /login
	15	+Disallow: /register
	16	+
	17	+# Landing pages reached from email links; nothing worth indexing.
	18	+Disallow: /preferences-saved
	19	+Disallow: /subscription-confirmed
	20	+
	21	+# The JSON API: no HTML, nothing to index.
	22	+Disallow: /api/
	23	+
	24	+Allow: /

tests/Unit/RobotsTxt.test.tsadded+22-0

Changes to tests/Unit/RobotsTxt.test.ts

		@@ -0,0 +1,22 @@
	1	+import { describe, expect, it } from 'bun:test'
	2	+import { readFileSync } from 'node:fs'
	3	+import { join } from 'node:path'
	4	+
	5	+const robots = readFileSync(join(import.meta.dir, '../../public/robots.txt'), 'utf8')
	6	+
	7	+// scripts/smoke-deployment.ts fetches /robots.txt on production after every
	8	+// deploy and fails the run if it is missing, names the wrong host, or leaks a
	9	+// localhost URL. Checking the same things here keeps that gate from going red
	10	+// only after the code is already live.
	11	+describe('robots.txt', () => {
	12	+ it('names the production host and no local one', () => {
	13	+ expect(robots).toContain('commshq.org')
	14	+ expect(robots).not.toContain('localhost')
	15	+ })
	16	+
	17	+ it('keeps crawlers out of the auth-walled and API routes', () => {
	18	+ for (const path of ['/dashboard', '/login', '/register', '/api/'])
	19	+ expect(robots).toContain(`Disallow: ${path}`)
	20	+ expect(robots).toContain('Allow: /')
	21	+ })
	22	+})