Add browserless FMHY site adapters

integrations/opencli/fmhy/README.md

@@ -0,0 +1,22 @@
+# FMHY OpenCLI adapter
+
+Public, browserless adapters for the live `https://fmhy.net/` sitemap. The adapter reads only URLs currently published by FMHY, checks `robots.txt`, limits concurrency, and returns structured page text and resource links.
+
+## Install
+
+```powershell
+& .\integrations\opencli\fmhy\install.ps1
+```
+
+Use `-Force` only to replace an existing managed FMHY adapter.
+
+## Commands
+
+```powershell
+opencli fmhy pages -f json
+opencli fmhy page ai -f json
+opencli fmhy crawl --group other -f json
+opencli fmhy search "open source" --limit 25 -f json
+```
+
+`crawl` discovers the current sitemap on each run. It therefore covers new or removed FMHY subpages without regenerating one command per URL.

integrations/opencli/fmhy/crawl.js

@@ -0,0 +1,28 @@
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import { crawlPages } from './utils.js';
+
+cli({
+    site: 'fmhy',
+    name: 'crawl',
+    access: 'read',
+    description: 'Crawl every live FMHY sitemap page with robots enforcement and bounded concurrency',
+    domain: 'fmhy.net',
+    strategy: Strategy.PUBLIC,
+    browser: false,
+    args: [
+        { name: 'group', type: 'str', required: false, help: 'Optional first path segment, such as posts or other' },
+        { name: 'max-pages', type: 'int', required: false, default: 0, help: 'Maximum pages to crawl; 0 means every matching sitemap page' },
+        { name: 'concurrency', type: 'int', required: false, default: 3, help: 'Concurrent requests, clamped to 1-6' },
+        { name: 'delay-ms', type: 'int', required: false, default: 150, help: 'Polite delay before each page request, capped at 2000ms' },
+    ],
+    columns: ['page', 'page_title', 'section', 'kind', 'title', 'description', 'url', 'links'],
+    func: async (kwargs) => {
+        const result = await crawlPages({
+            group: kwargs.group,
+            maxPages: kwargs['max-pages'],
+            concurrency: kwargs.concurrency,
+            delayMs: kwargs['delay-ms'],
+        });
+        return result.records;
+    },
+});

integrations/opencli/fmhy/install.ps1

@@ -0,0 +1,19 @@
+[CmdletBinding()]
+param(
+    [switch]$Force
+)
+
+$destination = Join-Path $env:USERPROFILE '.opencli\clis\fmhy'
+$adapterFiles = @('utils.js', 'pages.js', 'page.js', 'crawl.js', 'search.js')
+
+if ((Test-Path -LiteralPath $destination) -and -not $Force) {
+    throw "FMHY adapter already exists at $destination. Re-run with -Force to replace the managed files."
+}
+
+New-Item -ItemType Directory -Path $destination -Force | Out-Null
+foreach ($name in $adapterFiles) {
+    Copy-Item -LiteralPath (Join-Path $PSScriptRoot $name) -Destination (Join-Path $destination $name) -Force
+}
+
+Write-Output "Installed FMHY OpenCLI adapter to $destination"
+Write-Output 'Validate with: opencli validate fmhy'

integrations/opencli/fmhy/package.json

@@ -0,0 +1,7 @@
+{
+  "private": true,
+  "type": "module",
+  "scripts": {
+    "test": "node --test utils.test.mjs"
+  }
+}

integrations/opencli/fmhy/page.js

@@ -0,0 +1,17 @@
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import { fetchPage } from './utils.js';
+
+cli({
+    site: 'fmhy',
+    name: 'page',
+    access: 'read',
+    description: 'Fetch one live FMHY sitemap page as structured text and resource records',
+    domain: 'fmhy.net',
+    strategy: Strategy.PUBLIC,
+    browser: false,
+    args: [
+        { name: 'path', type: 'str', required: true, positional: true, help: 'FMHY path or full fmhy.net URL, for example ai or /other/selfhosting' },
+    ],
+    columns: ['page', 'page_title', 'section', 'kind', 'title', 'description', 'url', 'links'],
+    func: async (kwargs) => (await fetchPage(kwargs.path)).records,
+});

integrations/opencli/fmhy/pages.js

@@ -0,0 +1,21 @@
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import { fetchSitemap, sitemapRows } from './utils.js';
+
+cli({
+    site: 'fmhy',
+    name: 'pages',
+    access: 'read',
+    description: 'List every crawlable FMHY page from the live sitemap',
+    domain: 'fmhy.net',
+    strategy: Strategy.PUBLIC,
+    browser: false,
+    args: [
+        { name: 'group', type: 'str', required: false, help: 'Optional first path segment, such as posts or other' },
+    ],
+    columns: ['path', 'group', 'kind', 'url'],
+    func: async (kwargs) => {
+        const rows = sitemapRows(await fetchSitemap());
+        const group = String(kwargs.group ?? '').trim().toLowerCase();
+        return group ? rows.filter(row => row.group.toLowerCase() === group) : rows;
+    },
+});

integrations/opencli/fmhy/search.js

@@ -0,0 +1,38 @@
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import { crawlPages } from './utils.js';
+
+cli({
+    site: 'fmhy',
+    name: 'search',
+    access: 'read',
+    description: 'Search structured content across all live FMHY sitemap pages',
+    domain: 'fmhy.net',
+    strategy: Strategy.PUBLIC,
+    browser: false,
+    args: [
+        { name: 'query', type: 'str', required: true, positional: true, help: 'Case-insensitive text to search for' },
+        { name: 'group', type: 'str', required: false, help: 'Optional first path segment, such as posts or other' },
+        { name: 'limit', type: 'int', required: false, default: 50, help: 'Maximum matching records, capped at 500' },
+        { name: 'concurrency', type: 'int', required: false, default: 3, help: 'Concurrent requests, clamped to 1-6' },
+        { name: 'delay-ms', type: 'int', required: false, default: 150, help: 'Polite delay before each page request, capped at 2000ms' },
+    ],
+    columns: ['page', 'page_title', 'section', 'kind', 'title', 'description', 'url', 'links'],
+    func: async (kwargs) => {
+        const query = String(kwargs.query ?? '').trim().toLowerCase();
+        if (!query) throw new Error('FMHY search requires a non-empty query.');
+        const limit = Math.max(1, Math.min(Number(kwargs.limit) || 50, 500));
+        const result = await crawlPages({
+            group: kwargs.group,
+            concurrency: kwargs.concurrency,
+            delayMs: kwargs['delay-ms'],
+        });
+        return result.records.filter(record => [
+            record.page,
+            record.page_title,
+            record.section,
+            record.title,
+            record.description,
+            record.url,
+        ].some(value => String(value ?? '').toLowerCase().includes(query))).slice(0, limit);
+    },
+});