Open-source social media scheduler built for AI agents and the teams behind them. Create, schedule and publish posts to every network through MCP or the REST API, or write them yourself in a composer your team will love.
PHP ai-agentsblueskyfacebookinstagramlaravellinkedinmastodonmcpmcp-serveropen-sourcerest-apiself-hostedsocial-mediasocial-media-managementsocial-media-schedulerthreadstiktokvuejsxyoutube
<?php
declare(strict_types=1);
use Illuminate\Support\Str;
/*
* Every third-party URL goes through here, so a reader can append a path
* without trimming: a value set with a trailing slash comes back without it.
*/
$url = static fn (string $key, string $default): string => Str::rtrim((string) env($key, $default), '/');
return [
/*
|--------------------------------------------------------------------------
| Self-Hosted Mode
|--------------------------------------------------------------------------
|
| When enabled, the application runs in self-hosted mode which skips
| payment/subscription requirements during welcome.
|
*/
'self_hosted' => env('SELF_HOSTED', true),
/*
|--------------------------------------------------------------------------
| Changelog ("What's New")
|--------------------------------------------------------------------------
|
| Shows the sidebar card and the "What's New" modal fed by the entries
| under resources/changelog/, synced with `php artisan changelog:sync`
| on every deploy. On by default, self-hosted included.
|
*/
'changelog' => [
'enabled' => (bool) env('CHANGELOG_ENABLED', true),
'path' => resource_path('changelog'),
],
/*
|--------------------------------------------------------------------------
| In-app tours
|--------------------------------------------------------------------------
|
| The tours that walk a user through a page the first time they see it
| (`TourStep`). On by default; browser tests turn them off so a tour
| never covers the page they drive.
|
*/
'tours' => [
'enabled' => (bool) env('TOURS_ENABLED', true),
],
/*
|--------------------------------------------------------------------------
| Legal pages
|--------------------------------------------------------------------------
|
| Linked from the auth screens. Platform app reviews (TikTok explicitly)
| require Terms and Privacy links to be clearly visible; self-hosted
| installs point these at wherever they publish their own documents.
|
*/
'legal' => [
'terms_url' => $url('LEGAL_TERMS_URL', 'https://trypost.it/terms'),
'privacy_url' => $url('LEGAL_PRIVACY_URL', 'https://trypost.it/privacy'),
],
/*
|--------------------------------------------------------------------------
| Well-known files
|--------------------------------------------------------------------------
|
| Files served at /.well-known/{name} as plain text, keyed by name. The
| content always comes from the environment so no install serves another
| install's tokens; an empty value answers 404. Add a provider with one
| line here and one env variable.
|
*/
'well_known' => [
'openai-apps-challenge' => env('OPENAI_APPS_CHALLENGE'),
],
/*
|--------------------------------------------------------------------------
| ChatGPT MCP egress ranges
|--------------------------------------------------------------------------
|
| ChatGPT calls the MCP server from the ranges OpenAI publishes
| (https://developers.openai.com/api/docs/guides/ip-addresses). From
| there, a request without a token may list the tools, and a tool call
| is answered with the sign-in prompt ChatGPT shows in the conversation.
| Every other client keeps the plain 401.
|
*/
'mcp_chatgpt_ip_ranges_url' => $url('MCP_CHATGPT_IP_RANGES_URL', 'https://openai.com/chatgpt-connectors.json'),
/*
|--------------------------------------------------------------------------
| Claude MCP egress ranges
|--------------------------------------------------------------------------
|
| Claude calls MCP servers from the outbound ranges Anthropic publishes
| (https://platform.claude.com/docs/en/api/ip-addresses). Anthropic has
| no machine-readable list, so the ranges live here.
|
*/
'mcp_claude_ip_ranges' => explode(',', env('MCP_CLAUDE_IP_RANGES', '160.79.104.0/21')),
/*
|--------------------------------------------------------------------------
| MCP rate limit
|--------------------------------------------------------------------------
|
| Requests one MCP token (one AI client, for one user, in one workspace)
| may send to the MCP server. The REST API counts apart.
|
*/
'mcp_requests_per_minute' => (int) env('MCP_REQUESTS_PER_MINUTE', 240),
'mcp_requests_per_hour' => (int) env('MCP_REQUESTS_PER_HOUR', 2400),
/*
|--------------------------------------------------------------------------
| Meta page walk budget
|--------------------------------------------------------------------------
|
| Seconds the Facebook/Instagram page walk may spend before it returns what
| it has and reports itself incomplete. It runs inside the OAuth callback,
| so this must stay well under the web server's request timeout.
|
*/
'meta_page_walk_seconds' => (int) env('META_PAGE_WALK_SECONDS', 20),
/*
|--------------------------------------------------------------------------
| Security
|--------------------------------------------------------------------------
|
| SafeHttpFetcher blocks requests to private/reserved IP ranges (SSRF
| protection) by default. Self-hosted operators who need to fetch from
| their own internal network (e.g. an internal webhook endpoint) can
| opt in here. Leave disabled unless you understand the SSRF risk.
|
*/
'security' => [
'allow_private_network' => (bool) env('TRYPOST_ALLOW_PRIVATE_NETWORK', false),
],
/*
|--------------------------------------------------------------------------
| Media Size Limits
|--------------------------------------------------------------------------
|
| Per-type size caps in megabytes. Single source of truth — direct
| uploads (StoreChunkedMediaRequest), URL
| fetches (MediaAttacher), and the MediaType enum all read from here.
|
| signed_upload_url_ttl_minutes controls the temporary signed POST URL
| issued for api.uploads.store (MCP / direct upload flow).
| MEDIA_SIGNED_UPLOAD_URL_TTL_MINUTES is preferred; MCP_UPLOAD_URL_TTL_MINUTES
| remains as a legacy fallback. MCP_UPLOAD_MAX_SIZE_MB was removed — size
| caps come from max_size_mb above (uploads stream to storage).
|
| signed_upload_per_*_per_minute backs the signed-uploads rate limiter:
| workspace bucket first (tenants isolated on shared MCP egress), then a
| high IP backstop.
|
*/
'media' => [
'max_size_mb' => [
'image' => (int) env('MEDIA_IMAGE_MAX_SIZE_MB', 10),
'video' => (int) env('MEDIA_VIDEO_MAX_SIZE_MB', 1024),
// LinkedIn caps document (PDF carousel) uploads at 100MB.
'document' => (int) env('MEDIA_DOCUMENT_MAX_SIZE_MB', 100),
],
'image_max_pixels' => (int) env('MEDIA_IMAGE_MAX_PIXELS', 8192 * 8192),
'signed_upload_url_ttl_minutes' => (int) (env('MEDIA_SIGNED_UPLOAD_URL_TTL_MINUTES') ?? env('MCP_UPLOAD_URL_TTL_MINUTES', 15)),
'signed_upload_per_workspace_per_minute' => (int) env('MEDIA_SIGNED_UPLOAD_PER_WORKSPACE_PER_MINUTE', 60),
'signed_upload_per_ip_per_minute' => (int) env('MEDIA_SIGNED_UPLOAD_PER_IP_PER_MINUTE', 1200),
'upload_retention_hours' => (int) env('MEDIA_UPLOAD_RETENTION_HOURS', 24),
'heic_conversion' => (bool) env('MEDIA_HEIC_CONVERSION', true),
'heic_limits' => [
'memory_mb' => (int) env('MEDIA_HEIC_MEMORY_LIMIT_MB', 256),
'map_mb' => (int) env('MEDIA_HEIC_MAP_LIMIT_MB', 512),
'disk_mb' => (int) env('MEDIA_HEIC_DISK_LIMIT_MB', 1024),
'area_mb' => (int) env('MEDIA_HEIC_AREA_LIMIT_MB', 1024),
'width_px' => (int) env('MEDIA_HEIC_WIDTH_LIMIT_PX', 16384),
'height_px' => (int) env('MEDIA_HEIC_HEIGHT_LIMIT_PX', 16384),
],
],
/*
|--------------------------------------------------------------------------
| Composer media sources
|--------------------------------------------------------------------------
|
| A source shows in the composer's media menu only when its flag is on and
| its credentials (config/services.php) are set. Hosts and SDK URLs are
| overridable for self-hosters; download_hosts is a comma-separated list
| where `*.` matches any subdomain.
|
*/
'media_sources' => [
'giphy' => [
'enabled' => (bool) env('GIPHY_ENABLED', true),
'api' => 'https://api.giphy.com/v1',
'website' => 'https://giphy.com',
'download_hosts' => '*.giphy.com',
'rating' => 'g',
'requests_per_user_per_minute' => 30,
],
'unsplash' => [
'api' => $url('UNSPLASH_API_URL', 'https://api.unsplash.com'),
'website' => $url('UNSPLASH_WEBSITE_URL', 'https://unsplash.com'),
'download_hosts' => env('UNSPLASH_DOWNLOAD_HOSTS', 'images.unsplash.com'),
'requests_per_user_per_minute' => (int) env('UNSPLASH_REQUESTS_PER_MINUTE', 30),
],
'google_oauth' => [
'authorize_url' => $url('GOOGLE_MEDIA_AUTHORIZE_URL', 'https://accounts.google.com/o/oauth2/v2/auth'),
'token_api' => $url('GOOGLE_MEDIA_OAUTH_API', 'https://oauth2.googleapis.com'),
],
'google_drive' => [
'enabled' => (bool) env('GOOGLE_DRIVE_ENABLED', true),
'api' => $url('GOOGLE_DRIVE_API_URL', 'https://www.googleapis.com/drive/v3'),
'picker_sdk' => $url('GOOGLE_PICKER_SDK_URL', 'https://apis.google.com/js/api.js'),
'download_hosts' => env('GOOGLE_DRIVE_DOWNLOAD_HOSTS', 'www.googleapis.com,*.googleusercontent.com'),
],
'google_photos' => [
'enabled' => (bool) env('GOOGLE_PHOTOS_ENABLED', false),
'api' => $url('GOOGLE_PHOTOS_PICKER_API_URL', 'https://photospicker.googleapis.com'),
'download_hosts' => env('GOOGLE_PHOTOS_DOWNLOAD_HOSTS', 'lh3.googleusercontent.com,lh4.googleusercontent.com,lh5.googleusercontent.com,lh6.googleusercontent.com,video-downloads.googleusercontent.com'),
'polls_per_user_per_minute' => (int) env('GOOGLE_PHOTOS_POLLS_PER_MINUTE', 60),
'max_items' => (int) env('GOOGLE_PHOTOS_MAX_ITEMS', 10),
],
'canva' => [
'enabled' => (bool) env('CANVA_ENABLED', true),
'api' => $url('CANVA_API_URL', 'https://api.canva.com/rest/v1'),
'authorize_url' => $url('CANVA_AUTHORIZE_URL', 'https://www.canva.com/api/oauth/authorize'),
'code_challenge_method' => env('CANVA_CODE_CHALLENGE_METHOD', 's256'),
'download_hosts' => env('CANVA_DOWNLOAD_HOSTS', '*.canva.com'),
'editor_hosts' => env('CANVA_EDITOR_HOSTS', '*.canva.com'),
'export_timeout_seconds' => (int) env('CANVA_EXPORT_TIMEOUT_SECONDS', 120),
'export_poll_seconds' => (int) env('CANVA_EXPORT_POLL_SECONDS', 2),
],
'import_timeout_seconds' => (int) env('MEDIA_IMPORT_TIMEOUT_SECONDS', 600),
'imports_per_user_per_minute' => (int) env('MEDIA_IMPORTS_PER_MINUTE', 30),
],
/*
|--------------------------------------------------------------------------
| Post history retention
|--------------------------------------------------------------------------
|
| Published posts older than this many days are
| deleted with their media by posts:prune-history. Must be 1 or more.
|
*/
'posts' => [
'history_retention_days' => env('POST_HISTORY_RETENTION_DAYS', 730),
],
/*
|--------------------------------------------------------------------------
| External posts
|--------------------------------------------------------------------------
|
| Posts published directly on a network become posts in Sent. Only the
| newest `import_limit` publications of a channel are imported (0 turns
| importing off); on X only those from the last `x_import_days` days.
|
*/
'external_posts' => [
'import_limit' => (int) env('EXTERNAL_POSTS_IMPORT_LIMIT', 50),
'x_import_days' => (int) env('X_EXTERNAL_POSTS_IMPORT_DAYS', 30),
'match_window_minutes' => (int) env('EXTERNAL_POSTS_MATCH_WINDOW_MINUTES', 120),
'media_requests_per_network_per_minute' => (int) env('EXTERNAL_POSTS_MEDIA_REQUESTS_PER_MINUTE', 30),
],
/*
|--------------------------------------------------------------------------
| Analytics publication discovery
|--------------------------------------------------------------------------
|
| Discovery looks for new posts on every channel every
| `discovery_interval_hours` (X separately, since X bills each read), and
| reads back `discovery_overlap_hours` from the newest post it has seen.
|
*/
'analytics' => [
'discovery_interval_hours' => (int) env('PUBLICATION_DISCOVERY_INTERVAL_HOURS', 3),
'x_discovery_interval_hours' => (int) env('X_PUBLICATION_DISCOVERY_INTERVAL_HOURS', 24),
'discovery_overlap_hours' => (int) env('PUBLICATION_DISCOVERY_OVERLAP_HOURS', 6),
],
/*
|--------------------------------------------------------------------------
| Google Authentication
|--------------------------------------------------------------------------
|
| Enable or disable "Login with Google" on the login and register pages.
| Disable this if you don't have Google OAuth credentials configured.
|
*/
/*
|--------------------------------------------------------------------------
| Outbound User-Agent
|--------------------------------------------------------------------------
|
| Branded User-Agent applied to outbound HTTP from workspace webhooks so
| recipients know the request came from TryPost.it. Self-hosters can
| override it.
|
*/
'user_agent' => env('TRYPOST_USER_AGENT', 'TryPost.it/1.0 (+https://trypost.it)'),
/*
|--------------------------------------------------------------------------
| Repurpose
|--------------------------------------------------------------------------
|
| How often an active repurpose polls its source network for videos the
| workspace published outside TryPost. The scheduler ticks every five
| minutes and each repurpose is polled when it is due, so the interval is
| a runtime knob rather than a cron expression. Meta's Instagram quota is
| an app-wide pool (200 calls per hour per daily active user), so raise
| the interval before the pool tightens. `backoff_minutes` is used instead
| when the source answers with a rate-limit error.
|
*/
'repurpose' => [
'poll_interval_minutes' => (int) env('REPURPOSE_POLL_INTERVAL_MINUTES', 15),
'backoff_minutes' => (int) env('REPURPOSE_BACKOFF_MINUTES', 60),
],
/*
|--------------------------------------------------------------------------
| Favicons
|--------------------------------------------------------------------------
|
| Host of the favicon provider behind the internal `favicon/{domain}`
| route. The server fetches the icon so browsers never leak the domains
| they show to a third party.
|
*/
'favicon' => [
'api' => $url('FAVICON_API', 'https://icons.duckduckgo.com'),
],
/*
|--------------------------------------------------------------------------
| RSS Feeds
|--------------------------------------------------------------------------
|
| Create › Feeds. Each feed is refreshed every `poll_interval_minutes`;
| failures back off exponentially up to `max_backoff_minutes`. The
| directory is the curated "Explore feeds" list shown in the UI.
|
*/
'rss_feeds' => [
'poll_interval_minutes' => (int) env('RSS_FEEDS_POLL_INTERVAL_MINUTES', 480),
'max_backoff_minutes' => (int) env('RSS_FEEDS_MAX_BACKOFF_MINUTES', 1440),
'refresh_cooldown_minutes' => (int) env('RSS_FEEDS_REFRESH_COOLDOWN_MINUTES', 5),
'max_items_per_feed' => (int) env('RSS_FEEDS_MAX_ITEMS_PER_FEED', 100),
'max_feeds_per_workspace' => (int) env('RSS_FEEDS_MAX_PER_WORKSPACE', 50),
'og_image_max_items_per_poll' => (int) env('RSS_FEEDS_OG_IMAGE_MAX_ITEMS', 20),
'og_image_max_age_days' => (int) env('RSS_FEEDS_OG_IMAGE_MAX_AGE_DAYS', 14),
'og_image_max_page_bytes' => (int) env('RSS_FEEDS_OG_IMAGE_MAX_PAGE_BYTES', 256 * 1024),
'max_response_bytes' => (int) env('RSS_FEEDS_MAX_RESPONSE_BYTES', 5 * 1024 * 1024),
'fetch_budget_seconds' => (int) env('RSS_FEEDS_FETCH_BUDGET_SECONDS', 10),
'directory' => [
'favorites' => [
['name' => 'Creator Science', 'url' => 'https://creatorscience.com/rss/'],
['name' => 'Lindsey Gamble\'s Newsletter', 'url' => 'https://rss.beehiiv.com/feeds/3W8Cmlhot6.xml'],
['name' => 'Passionfruit', 'url' => 'https://passionfru.it/feed/'],
['name' => 'ICYMI by Lia Haberman', 'url' => 'https://liahaberman.substack.com/feed'],
['name' => 'Link in Bio', 'url' => 'https://www.milkkarten.net/feed'],
['name' => 'Geekout Newsletter', 'url' => 'https://rss.beehiiv.com/feeds/e2fyt8op5j.xml'],
['name' => 'SparkToro Blog', 'url' => 'https://sparktoro.com/blog/feed/'],
],
'tech' => [
['name' => 'Engadget', 'url' => 'https://www.engadget.com/rss.xml'],
['name' => 'The Verge', 'url' => 'https://www.theverge.com/rss/index.xml'],
['name' => 'Wired', 'url' => 'https://www.wired.com/feed/rss'],
['name' => 'TechCrunch', 'url' => 'https://techcrunch.com/feed/'],
['name' => 'Lifehacker', 'url' => 'https://lifehacker.com/feed/rss'],
['name' => 'Gizmodo', 'url' => 'https://gizmodo.com/feed'],
['name' => 'Mashable', 'url' => 'https://mashable.com/feeds/rss/all'],
['name' => 'Fast Company', 'url' => 'https://feeds.feedburner.com/fastcompany/headlines'],
['name' => 'Ars Technica', 'url' => 'https://feeds.arstechnica.com/arstechnica/features'],
['name' => 'The Keyword - Google', 'url' => 'https://blog.google/rss/'],
['name' => 'MacRumors', 'url' => 'https://feeds.macrumors.com/MacRumors-All'],
['name' => 'Android Central', 'url' => 'https://www.androidcentral.com/feed'],
['name' => 'TED Talks', 'url' => 'https://feeds.feedburner.com/tedtalks_video'],
['name' => 'Slashdot', 'url' => 'https://rss.slashdot.org/Slashdot/slashdotMain'],
['name' => 'VentureBeat', 'url' => 'https://feeds.feedburner.com/venturebeat/SZYF'],
['name' => 'TheNextWeb', 'url' => 'https://feeds2.feedburner.com/thenextweb'],
['name' => 'Hackaday', 'url' => 'https://hackaday.com/blog/feed'],
['name' => 'Make Magazine', 'url' => 'https://makezine.com/feed/'],
['name' => 'MIT Technology Review', 'url' => 'https://www.technologyreview.com/feed/'],
['name' => 'Hacker News', 'url' => 'https://news.ycombinator.com/rss'],
['name' => '9to5Mac', 'url' => 'https://9to5mac.com/feed/'],
['name' => 'Cnet', 'url' => 'https://www.cnet.com/rss/news/'],
['name' => 'Technology - The New York Times', 'url' => 'https://www.nytimes.com/svc/collections/v1/publish/https:/www.nytimes.com/section/technology/rss.xml'],
],
'news' => [
['name' => 'BBC News', 'url' => 'https://feeds.bbci.co.uk/news/rss.xml'],
['name' => 'The New York Times', 'url' => 'https://rss.nytimes.com/services/xml/rss/nyt/HomePage.xml'],
['name' => 'The Guardian', 'url' => 'https://www.theguardian.com/uk/rss'],
['name' => 'NPR', 'url' => 'https://feeds.npr.org/1001/rss.xml'],
['name' => 'The World This Week | The Economist', 'url' => 'https://www.economist.com/the-world-this-week/rss.xml'],
['name' => 'Time', 'url' => 'https://time.com/feed/'],
['name' => 'Slate Magazine', 'url' => 'https://slate.com/feeds/all.rss'],
['name' => 'The Atlantic', 'url' => 'https://www.theatlantic.com/feed/all/'],
['name' => 'Yahoo News', 'url' => 'https://news.yahoo.com/rss'],
['name' => 'ABC News', 'url' => 'https://abcnews.go.com/abcnews/topstories'],
['name' => 'Culture - The New Yorker', 'url' => 'https://www.newyorker.com/feed/culture'],
['name' => 'Vice', 'url' => 'https://www.vice.com/en/feed/'],
['name' => 'Vox', 'url' => 'https://www.vox.com/rss/index.xml'],
['name' => 'The Washington Post', 'url' => 'https://feeds.washingtonpost.com/rss/world'],
['name' => 'Al Jazeera', 'url' => 'https://www.aljazeera.com/xml/rss/all.xml'],
],
'business' => [
['name' => 'Business Insider', 'url' => 'https://feeds2.feedburner.com/businessinsider'],
['name' => 'Fast Company', 'url' => 'https://feeds.feedburner.com/fastcompany/headlines'],
['name' => 'Harvard Business Review', 'url' => 'https://hbr.org/resources/xml/atom/tip.xml'],
['name' => 'Seth\'s Blog', 'url' => 'https://seths.blog/feed/atom/'],
['name' => 'Entrepreneur', 'url' => 'https://www.entrepreneur.com/rss-feed/latest'],
['name' => 'The World This Week | The Economist', 'url' => 'https://www.economist.com/the-world-this-week/rss.xml'],
['name' => 'The Hubspot Marketing Blog', 'url' => 'https://blog.hubspot.com/marketing/rss.xml'],
['name' => 'Time', 'url' => 'https://time.com/feed/'],
['name' => 'VentureBeat', 'url' => 'https://feeds.feedburner.com/venturebeat/SZYF'],
['name' => 'Copyblogger', 'url' => 'https://copyblogger.com/feed/'],
['name' => 'Inc. Magazine', 'url' => 'https://www.inc.com/rss'],
['name' => 'Business - The New York Times', 'url' => 'https://www.nytimes.com/svc/collections/v1/publish/https:/www.nytimes.com/section/business/rss.xml'],
['name' => 'Tim Ferris', 'url' => 'https://tim.blog/feed/'],
['name' => 'Small Business Trends', 'url' => 'https://feeds.feedburner.com/SmallBusinessTrends'],
],
'art_media' => [
['name' => 'Design Milk', 'url' => 'https://design-milk.com/feed/'],
['name' => 'Colossal', 'url' => 'https://www.thisiscolossal.com/feed/'],
['name' => 'It\'s Nice That', 'url' => 'https://feeds2.feedburner.com/itsnicethat/SlXC'],
['name' => 'Booooooom', 'url' => 'https://www.booooooom.com/feed/'],
['name' => 'Contemporary Art Daily', 'url' => 'https://www.contemporaryartdaily.com/feed/'],
['name' => 'Post Secret', 'url' => 'https://postsecret.com/feed/'],
['name' => 'Ignant', 'url' => 'https://www.ignant.com/feed/'],
['name' => 'this isn\'t happiness', 'url' => 'https://feeds.feedburner.com/thisisnthappiness'],
['name' => 'Hyperallergic', 'url' => 'https://hyperallergic.com/feed/'],
['name' => 'ArtsJournal', 'url' => 'https://www.artsjournal.com/feed'],
['name' => 'Art and Design - The New York Times', 'url' => 'https://www.nytimes.com/svc/collections/v1/publish/https:/www.nytimes.com/section/arts/design/rss.xml'],
['name' => 'ARTNews', 'url' => 'https://www.artnews.com/feed/rss/'],
['name' => 'Hi-Fructose Magazine', 'url' => 'https://hifructose.com/feed/'],
['name' => 'Artsy', 'url' => 'https://www.artsy.net/rss/news'],
],
'entertainment' => [
['name' => 'Buzzfeed', 'url' => 'https://www.buzzfeed.com/index.xml'],
['name' => 'Rolling Stone', 'url' => 'https://www.rollingstone.com/feed/rss/'],
['name' => 'SlashFilm', 'url' => 'https://feeds.feedburner.com/slashfilm'],
['name' => 'IndieWire', 'url' => 'https://www.indiewire.com/feed/rss/'],
['name' => 'TMZ.com', 'url' => 'https://www.tmz.com/rss.xml'],
['name' => 'Cracked.com', 'url' => 'https://feeds.feedburner.com/CrackedRSS'],
['name' => 'Deadline.com', 'url' => 'https://deadline.com/feed/rss/'],
['name' => 'Movies - The New York Times', 'url' => 'https://www.nytimes.com/svc/collections/v1/publish/https:/www.nytimes.com/section/movies/rss.xml'],
['name' => 'Vulture', 'url' => 'https://feeds.feedburner.com/nymag/vulture'],
['name' => 'Variety', 'url' => 'https://variety.com/feed/rss/'],
['name' => 'Ain\'t It Cool News', 'url' => 'https://www.aintitcool.com/node/feed/'],
['name' => 'Nerdist', 'url' => 'https://nerdist.com/feed/'],
],
'science' => [
['name' => 'Wired', 'url' => 'https://www.wired.com/feed/rss'],
['name' => 'Scientific American', 'url' => 'https://www.scientificamerican.com/platform/syndication/rss/'],
['name' => 'MIT Technology Review', 'url' => 'https://www.technologyreview.com/feed/'],
['name' => 'New Scientist', 'url' => 'https://www.newscientist.com/feed/home/'],
['name' => 'ScienceDaily', 'url' => 'https://www.sciencedaily.com/rss/all.xml'],
['name' => 'Space.com', 'url' => 'https://www.space.com/feeds.xml'],
['name' => 'Science - The New York Times', 'url' => 'https://www.nytimes.com/svc/collections/v1/publish/https:/www.nytimes.com/section/science/rss.xml'],
['name' => 'Information is Beautiful', 'url' => 'https://feeds.feedburner.com/InformationIsBeautiful'],
['name' => 'Futurity', 'url' => 'https://www.futurity.org/feed/'],
['name' => 'Phys.org', 'url' => 'https://phys.org/rss-feed/'],
['name' => 'Futurism', 'url' => 'https://futurism.com/feed'],
['name' => 'Science and Technology | The Economist', 'url' => 'https://www.economist.com/science-and-technology/rss.xml'],
],
],
],
'google_auth_enabled' => env('GOOGLE_AUTH_ENABLED', false),
'github_auth_enabled' => env('GITHUB_AUTH_ENABLED', false),
/*
|--------------------------------------------------------------------------
| Social Platforms
|--------------------------------------------------------------------------
|
| Configure which social platforms are enabled in the application.
| Set to false to temporarily disable a platform (e.g., when credentials
| are revoked, expired, or pending approval).
|
*/
'platforms' => [
'linkedin' => [
'enabled' => env('LINKEDIN_ENABLED', true),
'api' => $url('LINKEDIN_API', 'https://api.linkedin.com'),
// OAuth host is different from the data API (api.linkedin.com).
'oauth_api' => $url('LINKEDIN_OAUTH_API', 'https://www.linkedin.com'),
// Scopes for LinkedIn authentication
'scopes' => array_values(array_filter(array_map('trim', explode(',', (string) env('LINKEDIN_SCOPES', 'openid,profile,email,w_member_social'))))),
],
'linkedin-page' => [
'enabled' => env('LINKEDIN_PAGE_ENABLED', true),
'api' => $url('LINKEDIN_PAGE_API', 'https://api.linkedin.com'),
// Scopes for LinkedIn Page authentication
'scopes' => array_values(array_filter(array_map('trim', explode(',', (string) env('LINKEDIN_PAGE_SCOPES', 'openid,profile,email,w_organization_social,r_organization_social,rw_organization_admin,w_member_social'))))),
],
'x' => [
'enabled' => env('X_ENABLED', true),
'api' => $url('X_API', 'https://api.x.com/2'),
'defuse_links' => (bool) env('X_DEFUSE_LINKS', false),
],
'tiktok' => [
'enabled' => env('TIKTOK_ENABLED', true),
'api' => $url('TIKTOK_API', 'https://open.tiktokapis.com/v2'),
// OAuth scopes to request. Trim when the TikTok app lacks a product
// (e.g. no Display API => drop user.info.profile, user.info.stats,
// video.list; analytics degrade gracefully, username stays empty).
'scopes' => array_values(array_filter(array_map('trim', explode(',', (string) env('TIKTOK_SCOPES', 'user.info.basic,user.info.profile,user.info.stats,video.publish,video.upload,video.list'))))),
],
'youtube' => [
'enabled' => env('YOUTUBE_ENABLED', true),
'data_api' => $url('YOUTUBE_DATA_API', 'https://www.googleapis.com/youtube/v3'),
'analytics_api' => $url('YOUTUBE_ANALYTICS_API', 'https://youtubeanalytics.googleapis.com/v2'),
'oauth_api' => $url('YOUTUBE_OAUTH_API', 'https://oauth2.googleapis.com'),
],
'facebook' => [
'enabled' => env('FACEBOOK_ENABLED', true),
'graph_api' => $url('FACEBOOK_GRAPH_API', 'https://graph.facebook.com/v25.0'),
'rupload_host' => env('FACEBOOK_RUPLOAD_HOST', 'rupload.facebook.com'),
],
'instagram' => [
'enabled' => env('INSTAGRAM_ENABLED', true),
'graph_api' => $url('INSTAGRAM_GRAPH_API', 'https://graph.instagram.com/v25.0'),
// graph.instagram.com (no version) is the auth/refresh host.
'auth_api' => $url('INSTAGRAM_AUTH_API', 'https://graph.instagram.com'),
'oauth_url' => $url('INSTAGRAM_OAUTH_URL', 'https://www.instagram.com'),
'oauth_api' => $url('INSTAGRAM_OAUTH_API', 'https://api.instagram.com'),
],
'instagram-facebook' => [
'enabled' => env('INSTAGRAM_FACEBOOK_ENABLED', true),
'graph_api' => $url('INSTAGRAM_FACEBOOK_GRAPH_API', 'https://graph.facebook.com/v25.0'),
],
'threads' => [
'enabled' => env('THREADS_ENABLED', true),
'graph_api' => $url('THREADS_GRAPH_API', 'https://graph.threads.net/v1.0'),
// graph.threads.net (no version) is the auth/refresh host.
'auth_api' => $url('THREADS_AUTH_API', 'https://graph.threads.net'),
'oauth_url' => $url('THREADS_OAUTH_URL', 'https://threads.net'),
],
'pinterest' => [
'enabled' => env('PINTEREST_ENABLED', true),
'api' => $url('PINTEREST_API', 'https://api.pinterest.com/v5'),
],
'bluesky' => [
'enabled' => env('BLUESKY_ENABLED', true),
'public_appview' => $url('BLUESKY_PUBLIC_APPVIEW', 'https://public.api.bsky.app'),
// Default PDS used when the account has no `meta.service` override.
'default_service' => $url('BLUESKY_DEFAULT_SERVICE', 'https://bsky.social'),
// Web client where published posts are viewed (profile/post URLs).
'web_app' => $url('BLUESKY_WEB_APP', 'https://bsky.app'),
// Video upload service (separate from the PDS). Videos are processed
// here, then the resulting blob is embedded in the post record.
'video_service' => $url('BLUESKY_VIDEO_SERVICE', 'https://video.bsky.app'),
'video_service_did' => env('BLUESKY_VIDEO_SERVICE_DID', 'did:web:video.bsky.app'),
// Seconds between transcode job-status polls.
'video_poll_seconds' => env('BLUESKY_VIDEO_POLL_SECONDS', 2),
// Gradually back off status checks to at most this interval.
'video_poll_max_seconds' => env('BLUESKY_VIDEO_POLL_MAX_SECONDS', 30),
// Bluesky rejects videos over 300 MB (app.bsky.embed.video maxSize = 300000000).
'video_max_bytes' => env('BLUESKY_VIDEO_MAX_BYTES', 300_000_000),
// PLC directory, used to resolve an account's real PDS host from its DID.
'plc_directory' => $url('BLUESKY_PLC_DIRECTORY', 'https://plc.directory'),
],
'mastodon' => [
'enabled' => env('MASTODON_ENABLED', true),
// Default instance used when the account has no `meta.instance` override.
'default_instance' => $url('MASTODON_DEFAULT_INSTANCE', 'https://mastodon.social'),
],
'telegram' => [
'enabled' => env('TELEGRAM_ENABLED', true),
// Single shared bot (BotFather). Users add it as admin to their channel.
'bot_token' => env('TELEGRAM_BOT_TOKEN'),
'bot_username' => env('TELEGRAM_BOT_USERNAME'),
'deep_link' => $url('TELEGRAM_DEEP_LINK', 'https://t.me'),
'api' => $url('TELEGRAM_API', 'https://api.telegram.org'),
// Secret-token header Telegram echoes on every webhook call.
'webhook_secret' => env('TELEGRAM_WEBHOOK_SECRET'),
],
'discord' => [
'enabled' => env('DISCORD_ENABLED', true),
// Single shared bot application. OAuth (bot scope) authorizes adding the
// bot to the user's server; channel listing, mentions and posting all
// use this bot token, not the user's OAuth token.
'bot_token' => env('DISCORD_BOT_TOKEN'),
'api' => $url('DISCORD_API', 'https://discord.com/api/v10'),
'oauth_api' => $url('DISCORD_OAUTH_API', 'https://discord.com/api/oauth2'),
// Permission bitfield requested for the bot: VIEW_CHANNEL (1<<10) +
// SEND_MESSAGES (1<<11) + EMBED_LINKS (1<<14) + ATTACH_FILES (1<<15) +
// READ_MESSAGE_HISTORY (1<<16) + MENTION_EVERYONE (1<<17) = 248832.
'permissions' => env('DISCORD_PERMISSIONS', '248832'),
'scopes' => array_values(array_filter(array_map('trim', explode(',', (string) env('DISCORD_SCOPES', 'bot,identify,guilds'))))),
],
'google_business' => [
'enabled' => env('GOOGLE_BUSINESS_ENABLED', true),
// Account Management API — lists the Business accounts a user administers.
'account_management_api' => $url('GOOGLE_BUSINESS_ACCOUNT_MANAGEMENT_API', 'https://mybusinessaccountmanagement.googleapis.com/v1'),
// Business Information API — lists locations under an account.
'business_information_api' => $url('GOOGLE_BUSINESS_BUSINESS_INFORMATION_API', 'https://mybusinessbusinessinformation.googleapis.com/v1'),
// Legacy but still-active v4 API — the only home for Local Post create/update/delete.
'local_posts_api' => $url('GOOGLE_BUSINESS_LOCAL_POSTS_API', 'https://mybusiness.googleapis.com/v4'),
// OAuth token endpoint, same host Google uses for every OAuth2 client.
'oauth_api' => $url('GOOGLE_BUSINESS_OAUTH_API', 'https://oauth2.googleapis.com'),
// Business Profile web UI — post URL fallback and the social-account profile link.
'dashboard' => $url('GOOGLE_BUSINESS_DASHBOARD', 'https://business.google.com'),
],
],
];