* Fix OpenRouter support by using laravel/ai default provider config.
PR #216 patched Lab::OpenRouter into every agent match, but laravel/ai
already resolves config('ai.default') — including openrouter — when agents
omit provider(). Those matches also forced unknown providers to Gemini and
BrandAnalyzerRunner checked a non-existent services.openrouter key.
Remove the duplicated provider() overrides, gate availability on
ai.providers.*.key, and document OPENROUTER_API_KEY.
Co-authored-by: Paulo Castellano <hello@paulocastellano.com>
* chore(deps): bump laravel/ai to v0.10.3 for native OpenRouter support
v0.5.1's OpenRouter driver only covered text/embeddings via the legacy
Prism gateway. v0.10.3 ships a native OpenRouter gateway with image,
audio (TTS/STT), and web search support, and drops prism-php/prism as
a dependency. ai-sdk-development skill docs refreshed via boost:update
to match the installed version.
* fix: honor AI_IMAGE_PROVIDER instead of hardcoding OpenAI's gpt-image-2
AiImageClient always passed model: 'gpt-image-2' to Image::of()->generate(),
so AI_IMAGE_PROVIDER silently did nothing for any provider other than
OpenAI — gemini/xai/openrouter would fail against an OpenAI-only model id
and quietly fall back to a stock photo. Usage recording was hardcoded to
provider 'openai' too, so credits were billed against the wrong model
whenever a different provider actually ran.
Drop the hardcoded model so generation falls through to the SDK's own
config('ai.default_for_images') + per-provider default model, and read
the actual provider/model back off the response's meta for usage
recording and source_meta instead of assuming OpenAI.
* refactor: extract AiImageClient into single-purpose steps, fix uncaught exception
generate() built the prompt, called the SDK, and unpacked the response all
in one block, with bytes extraction happening after the try/catch — so a
response with an empty images collection threw an uncaught RuntimeException
from ImageResponse::firstImage() instead of returning null as documented.
Split into cleanKeywords(), buildPrompt(), resolveBrandContext(), and
toResult(), and moved response unpacking inside the try block so any
malformed response is treated as a failure like everything else. Added a
regression test with an empty-images fake response.
* fix: drop hardcoded default_text_model, resolve model per provider
default_text_model was the only per-modality model override in ai.php —
image, audio, transcription, embeddings, and reranking all just pick a
provider and let it use its own default model. Text had a config-pinned
model on top, forced into every agent's model() and into every usage
log's model field regardless of which provider actually ran. Switching
AI_TEXT_PROVIDER (e.g. to openrouter) kept sending OpenAI's model id to
whichever provider ended up handling the request.
Removed model() from all six agents so laravel/ai resolves the model
from the active provider's own default (OpenAI's default is already
'gpt-5.4', so no behavior change there). Usage-recording call sites now
read the actual provider/model back off the response's meta instead of
assuming config('ai.default')/default_text_model. StreamPostContent
needed the then() callback since broadcast()'s StreamableAgentResponse
doesn't expose meta directly.
* refactor: drop AiConfiguration wrapper, use data_get() for array reads
AiConfiguration was a one-line static helper used by only two call
sites, with no laravel/ai equivalent to lean on (confirmed AiManager
and the Provider base class expose no isConfigured()/hasKey() check —
the package's model is try-then-catch, not pre-flight checks). Inlined
the filled(config(...)) check directly into HandleInertiaRequests and
BrandAnalyzerRunner instead of keeping a class around one line of logic.
Also swapped direct array-key reads for data_get() per project
convention across every file touched by the recent AI provider/model
fixes (agents' budget arrays, RunGenerateNode/StreamPostCreation's
humanizer merge, RegeneratePostMediaImage's baseContext/copy/rendered
access). Write/assignment sites (`$x['key'] = ...`) are left as-is —
data_get() only reads.
* feat: enhance AI configuration with new providers and options
Added strict types declaration and updated Azure OpenAI API version. Introduced new 'bedrock' provider configuration with AWS credentials and role assumptions. Enhanced existing providers with additional options, including image deployment for Azure and OpenAI, and updated URLs for Gemini and Ollama. Added support for an 'openai-compatible' driver to broaden integration capabilities.
* fix: remove trailing newline in AI configuration file
* feat: add support for OpenRouter and ElevenLabs API keys in configuration
Updated the production Docker Compose file and example environment file to include commented-out entries for OPENROUTER_API_KEY and ELEVENLABS_API_KEY. This enhances the configuration options for AI providers, allowing for easier integration of additional services.
* feat: allow per-provider model overrides for every AI modality
Adds a `models` array to each provider block, wired to env vars, so
self-hosted operators can pin a specific text/image/audio/transcription/
embeddings/reranking model instead of relying on the package's built-in
default for that provider. Only added for the modalities each provider
actually implements (verified against laravel/ai's Provider classes).
Azure is left untouched — it resolves models via deployment names
(AZURE_OPENAI_DEPLOYMENT etc.), not raw model strings, which was already
wired before this change.
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* fix: subscribe to AI generation channel before dispatching the job
AiGenerateDialog posted the generate request and only subscribed to the
broadcast channel once the response came back. Reverb's plain Channel
(vendor/laravel/reverb/.../Channels/Channel.php) delivers broadcasts by
iterating the in-memory connection list at send time only — no history,
no replay. If the queued job started streaming text_delta/stream_end
events before the private-channel subscribe handshake finished, those
events were gone for good and the dialog hung on 'streaming' forever.
Generate the generation_id client-side, await channel subscription
confirmation (Echo's channel.subscribed(), backed by
pusher:subscription_succeeded) before sending the generate request, and
have the backend use the client-supplied id instead of minting its own.
Related to #218.
* fix: handle subscribe failure, dialog-close race, and i18n hardcoded strings
Code review on the previous commit surfaced real gaps:
- useAiStream only handled the subscribe-success path (.subscribed()).
A definitive pusher:subscription_error (expired session, CSRF
mismatch) fell through to the 5s timeout and was treated as success,
reproducing the exact #218 hang via a different trigger. subscribe()
now returns a boolean: true on confirm-or-ambiguous-timeout (proceed
optimistically), false only on an explicit error (don't dispatch work
nothing will ever hear).
- AiGenerateDialog now bails without dispatching if the dialog is
closed while awaiting subscription confirmation (up to 5s), and
unsubscribes on a post-subscribe dispatch failure so stale listeners
can't flip status to 'completed' after an error was already shown.
- Extracted aiGenerationChannel() so the frontend has one place to
update if the channel format changes, and cross-referenced the two
remaining backend copies (PostAiGenerateController, StreamPostContent).
- Replaced the two hardcoded English fallback error strings with i18n
keys (posts.ai.generate.errors.*) across all 16 locales — this app
only ships translated copy.
- Simplified subscribe(): dropped a redundant "already settled" guard
and the clearTimeout bookkeeping — Promise.resolve() already no-ops
after the first call, so there was nothing to guard.
* fix: subscribe before dispatch in AI post creation and image regeneration flows
Extracts subscribePrivateChannel as a shared helper and applies the same
subscribe-before-dispatch fix from the post edit flow (issue #218) to the
other two flows that stream over private channels.
Post creation wizard moves the StreamPostCreation dispatch out of
AiPostWizard.vue and into Loading.vue, so the channel is subscribed before
the job that broadcasts on it is dispatched — previously a full page
navigation happened between dispatch and subscribe with no timeout
fallback, so a lost event hung the page forever.
* refactor: trim explanatory comments added in the subscribe-before-dispatch fix
Rationale lives in the commit history, not scattered inline comments.
* refactor: trim remaining explanatory comments from the subscribe-before-dispatch fix
Same cleanup as the previous commit, applied to the files from the
original post-edit-flow fix.
* fix: dedup post-creation dispatch, surface field-level AI errors, harden edge cases
Uses ShouldBeUnique on StreamPostCreation (scoped per-user) instead of an
ad-hoc cache guard, matching the pattern already used by
PublishToSocialPlatform and VerifyUpcomingPostConnections.
Field-level validation errors (prompt/instruction) now render inline via
InputError, matching how every other form in the app surfaces them, instead
of a generic status banner. Also fixes a 422 response missing the `errors`
key that Inertia's client silently swallows, adds error handling around the
wizard's navigation to the loading page, and extracts the duplicated
error-message parsing into a shared helper.
* Default AI content language to Ukrainian
* Keep English as the default content language
* refactor: update content language handling in workspace creation
- Changed the default content language in CreateWorkspace to inherit the app's locale instead of defaulting to English.
- Updated related tests to reflect this change, ensuring that the content language aligns with the application's current locale settings.
- Cleaned up unnecessary code in the BrandTab component for better readability.
* Add Ukrainian as a full platform UI locale.
Wire uk into languages config with complete lang/uk translations and restore ContentLanguage↔UI parity checks.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Wire Ukrainian into dayjs and locale coverage tests.
Import the uk dayjs locale with Monday week-start and cover uk/uk-UA in brand autofill and SetLocale assertions.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Harden Ukrainian UI and AI content-language coverage.
Cover generator/reviewer/humanizer/image prompts, workspace pickers, persona labels, UI locale switch, and README for uk as a first-class platform language.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Add Ukrainian Pinterest form strings after main merge.
Restore lang/uk posts.php key parity for the new pin title and destination link fields.
Co-authored-by: Cursor <cursoragent@cursor.com>
* List every supported UI language in the README.
Replace the abbreviated multi-language blurb with the full locale set from config.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Paulo Castellano <paulo@castellanos.llc>
Co-authored-by: Cursor <cursoragent@cursor.com>
Reject CreatePost calls without CreatedVia, set Web on DuplicatePost, and assert wiring for templates, AI, automation, and API spoof attempts.
Co-authored-by: Cursor <cursoragent@cursor.com>
Resolves the AiPostWizard conflict and completes the feature:
- i18n parity: brand_colors_label + brand_colors_description in all 15 locales
(was en/es/pt-BR only, which broke LocalizationParityTest).
- Reworked the two-button toggle into a Switch with an explanatory description
(matches the settings Switch/card pattern).
- Only shown for templates that honor the flag: added appliesBrandVisuals() to
the AiContentTemplate contract (ImageCard=true, tweet cards=false), exposed as
applies_brand_visuals in the create-page DTO, and gated the toggle on it — so
it no longer appears (as a no-op) for tweet-card styles.
- Tests: TemplateContractTest covers appliesBrandVisuals for all templates.
- Assert GeneratePostContentRequest rejects a prompt over the shared max, so the
editor's limit is pinned explicitly (not only implied by the create wizard).
- Assert the create wizard accepts a prompt at exactly the minimum length,
complementing the below-minimum rejection.
The counter added earlier drifted from the backend in two ways: it counted
UTF-16 code units over the raw (untrimmed) value, while the backend measures
Unicode characters (mb_strlen) over the trimmed value that is actually sent —
so emoji or trailing whitespace could falsely turn the counter red and block
the button. The 2000 limit was also copied into three places, and the wizard's
frontend `>= 3` minimum had no backend counterpart.
- Add App\Support\AiPromptRules as the single source of truth for the prompt
bounds; both StartPostCreationRequest and GeneratePostContentRequest use it.
- Add min:3 to the create wizard endpoint so front and back agree (the editor's
generate-content flow keeps `required` — it has no counter to mirror).
- Count code points over the trimmed value in AiPostWizard so the counter and
the submit gate match what the backend validates, matching AltTextDialog.
- Cover min/max/boundary in PostAiCreateTest.
The regenerate job replaced the media item without carrying its meta, silently dropping the user's alt text (and slide metadata) from the persisted post.
Copy meta from the freshly-locked post row (not the pre-render snapshot) inside the transaction, so an alt edit made while the multi-second render runs is kept rather than overwritten.
- Extract ContentLanguageOption into @/types and use it for availableContentLanguages
across BrandForm, BrandTab, Brand, Create, and the LanguagePicker options, so the
englishName field that drives search is visible to TypeScript instead of being
dropped silently by the pass-through prop types.
- Add BrandAnalyzerTest: assert the language schema enum equals the full 15-language
set (guards against it shrinking back to a hardcoded subset behind ::fake()) and
that instructions() lists every code.
- Isolate the "LLM language wins" autofill test by declaring the page as `en` while
the LLM returns `de`, so it actually proves mergeLlm precedence instead of both
paths agreeing.
- Cover SetLocale's cookie side effect: the default locale cookie is set on an
invalid/absent cookie and left untouched for a valid locale.
- Brand-analyzer prompt now lists every supported language instead of only
en/pt-BR/es, so onboarding autofill can detect the 12 added languages. The
backtick-formatted list is built in BrandAnalyzer::instructions(), keeping the
Blade clean and the enum free of prompt presentation.
- Translate the delete-confirmation keyword for el/ja/zh/ar (the four locales
that still shipped the English "delete").
- Make the language and font comboboxes RTL-correct (logical ms-* instead of
physical ml-*), and let the language combobox be searched by English name via
a visually-hidden label (ContentLanguage::options() now exposes englishName).
- Correct the ContentLanguage class docblock: the enum is also the source of
truth for the UI locales' text direction.
Tests: SetLocale middleware dir/RTL, isRtl and the full 15-language
englishName/label match arms, LLM language detection beyond en/es/pt-BR, and
store-path persistence of a non-default content language plus rejection of an
unsupported one.
Extend the homepage-language detection dataset to assert all 15 supported
languages resolve from their <html lang> subtag, and pin the primary-subtag
matching so a malformed tag ("english") no longer resolves via a two-letter
prefix.
The new content-language options were hand-duplicated across request
validation, the UI picker, and homepage detection, while the brand
analyzer's structured-output enum and the AI image prompt's language
name still only knew about en/pt-BR/es. That left autofill unable to
detect the new languages and made image text fall back to English for
them.
Introduce App\Enums\Workspace\ContentLanguage as the single source of
truth and derive every site from it:
- Store/UpdateWorkspaceRequest validate against ContentLanguage::values()
- BrandAnalyzer's language enum uses ContentLanguage::values()
- AiImageClient::languageName() resolves via the enum's englishName()
- HomepageMetaExtractor detects through ContentLanguage::fromHtmlLang()
- BrandForm consumes availableContentLanguages from the backend, like
availableFonts/availableImageStyles, instead of a hardcoded list
Also fix two labels: nl "Nederlandse" -> "Nederlands", zh -> "中文".
The image pipeline already threads `applyBrandVisuals` through
`TemplateContext` -> `PostImagePipeline` -> `TemplateImageGenerator`, but it was
hardcoded to `true` at the dispatch site, so generated images always used the
workspace brand palette with no way to opt out.
Expose the choice in the create wizard: a "Brand colors" / "Let AI decide"
toggle (shown only when images are generated). The flag flows
front -> `StartPostCreationRequest` (`apply_brand_visuals`) ->
`PostAiCreateController@start` -> `StreamPostCreation` -> `TemplateContext`,
defaulting to `true` so existing behavior is unchanged when the field is absent.
Replace the 'single'/'carousel' string format threaded through the AI
generation path with a typed GeneratorFormat enum (PostContentGenerator,
PostContentHumanizer, PostContentStreamer, StreamPostCreation,
RunGenerateNode, deriveFormat).
Remove generatorFormat() from AiContentTemplate and the three templates —
it conflated style and format (returning 'tweet_card'). The tweet-card
check now lives on ContentStyle via isTweetCard(), and the humanizer skip
on ContentStyle::humanizes(), so the style declares its own behaviour
instead of callers comparing magic strings.
- TemplateContext gains bool $isCarousel (set from format === CAROUSEL_FORMAT in StreamPostCreation)
- AiContentTemplate interface: promptView() now takes TemplateContext
- ImageCardTemplate handles both single and carousel via context.isCarousel; schema() returns the correct shape per mode; assemble() routes to assembleSingle/assembleCarousel
- TweetCardTemplate handles both single and carousel; promptView() returns tweet_card_carousel view when isCarousel; schema() returns {tweet_text} or {caption, slides[]}; assemble() calls forTweetCard (single) or forTweetCardCarousel (carousel)
- CarouselTemplate deleted — carousel is now a format dimension each style handles, not a standalone style
- AiTemplateRegistry: [ImageCardTemplate, TweetCardTemplate] only
- PostImagePipeline: adds forTweetCardCarousel — renders one renderTweetCard per slide text, returns N media items
- StreamPostCreation: carousel detection restored to $format === CAROUSEL_FORMAT (like main); style dispatched via AiTemplateRegistry::find($this->template); humanize skips tweet_card (single and carousel)
- PostContentGenerator: promptView() call updated to pass templateContext
- New prompt: resources/views/prompts/post_content/tweet_card_carousel.blade.php
- Tests: TemplateContractTest and TweetCardTemplateTest updated for promptView(context) signature; AiTemplateRegistryTest asserts [image_card, tweet_card]; StreamPostCreationTest adds tweet_card carousel test; TweetCardTemplateTest covers both schema shapes and both promptView return values
- Add AiTemplateRegistry (resolves, keys, default)
- Add template/templateContext params to PostContentGenerator; schema() delegates to template when both are set; instructions() uses template.promptView()
- Refactor StreamPostCreation.handle() to resolve template via registry, build TemplateContext, delegate assemble() to template; add createPostFromGenerated()
- Remove handleSingle(), handleCarousel(), resolvedContentType() (no other callers)
- PostAiCreateController passes template param from request (default image_card)
- All existing AI tests green; registry test added
Replace free-text brand_tone/brand_voice_notes with a single structured
brand_voice_traits JSON column backed by the BrandVoiceTrait enum, exposed
as choice-chip pills in the brand settings UI and autofillable from a site.
Brand voice and visuals become per-automation toggles on the Generate node.
Unify the image controls into one 0-10 picker (0 = text-only, 1 = single,
2+ = carousel) and feed the generator the most restrictive selected network
so copy fits every platform. Pass that same platform context through the
humanizer pass — extracted into a shared ResolvesPlatformCopyBudget trait —
so the rewrite can no longer drift past the character cap the generator
respected, in both the automation and manual creation flows.
Persist the trigger node's schedule editor fields on save (they were
silently dropped by validated() for lacking validation rules).
Instagram carousels were stored as content_type=instagram_carousel, which the
publisher's match() did not handle — publishing failed with "Unsupported
Instagram content type: instagram_carousel" for any post created via API, MCP,
or template (the AI flow worked only thanks to an inline band-aid that rewrote
carousel to feed before saving).
A carousel is just an Instagram feed post with multiple images: the editor,
preview, and publisher already treat a multi-image feed as a carousel. So
instagram_carousel is a generation format, not a stored content type. Remove it
from the ContentType enum entirely; it now lives only as an AI generation-format
string (wizard card + slide structure + carousel templates are untouched), and
posts always persist as instagram_feed.
- ContentType: drop the InstagramCarousel case; InstagramFeed maxMediaCount 1 -> 10
- StreamPostCreation: resolvedContentType() maps carousel -> feed; band-aid removed
- StartPostCreationRequest: accept instagram_carousel as a generation format
- Frontend: carousel becomes a wizard-local AiFormat; UI/UX unchanged
- API/MCP now reject instagram_carousel as a content_type (Rule::in no longer lists it)
- Replaced the `regenerate_image` boolean with a `change_mode` string parameter in the PostImageRegenerator schema, allowing for more granular control over regeneration options.
- Updated methods in RegeneratePostMediaImage to handle the new `change_mode` values: `image_only`, `text_only`, and `both`.
- Enhanced documentation and comments to clarify the new regeneration logic.
- Added tests to verify correct behavior for different regeneration modes, ensuring that text and image changes are handled appropriately.
- Replaced direct status checks in multiple controllers and actions with the PostStatusGuard utility, improving code readability and maintainability.
- Updated error messages to utilize a centralized method for consistency across the application.
- Removed the BrandImagePalette class, consolidating color resolution logic into the AiImageClient for better organization and type safety.
Let users adjust AI-generated slides in place via async job and Echo, while applying workspace brand, background, and text colors to image prompts. Autofill swaps site text/background colors for the image palette, and regeneration is blocked on finalized posts with safer job cleanup.
Co-authored-by: Cursor <cursoragent@cursor.com>
The whole app/Ai/PlatformRules/ system was registered in AppServiceProvider
but never read in production code. Confirmed via grep across app/, resources/,
config/, database/, tests/ — only consumers were:
- AppServiceProvider::configurePlatformRules() registering everything in the
Registry (which nothing then called)
- tests/Feature/Ai/PlatformRules/RegistryTest.php asserting the registrations
it had just performed
The actual AI text agent (PostContentGenerator) reads the per-platform caps
straight from the Platform enum (maxContentLength / recommendedAiContentLength)
and feeds them into the prompt template — bypassing this layer entirely.
Removed:
- app/Ai/PlatformRules/ (12 files: Contract, Registry, 10 Rule classes)
- tests/Feature/Ai/PlatformRules/RegistryTest.php
- 11 use imports + boot() call + configurePlatformRules() method in
AppServiceProvider
Core changes:
- Replace Unsplash slide pipeline with gpt-image-2 via Laravel AI SDK.
New AiImageClient builds prompts from a Blade template seeded by the
workspace's ImageStyle enum, content language, brand color (mapped to a
human-readable name via HexColorName helper) and brand description.
- Drop Template B from TemplateImageGenerator: every slide now renders as
Template A (full-bleed photo + bottom gradient + white/grey overlay).
Removes renderTemplateB, roundCorners, blendHex, ensureContrast and the
closing-slide pipeline.
- StreamPostCreation creates the Post directly and dispatches
PostCreationReady with post_id; the wizard kills its preview step and
redirects straight to the post editor on completion. Finalize endpoint
removed.
- New Workspace.image_style enum field with an 8-option visual picker
shared by /workspaces/create and /settings/workspace/brand via a single
BrandForm component (autofill is a prop). 8 sample webp thumbs ship
under public/images/branding/image-styles/.
- Media items gain optional source ('ai'|'unsplash'|'giphy') and
source_meta (recipe needed to regenerate AI images later); the gallery
picker tags Unsplash/Giphy attachments.
- Brand-color autofill: new CssColorFrequencyExtractor parses every
hex/rgb/hsl value in the homepage CSS, clusters perceptually similar
shades in CIE LAB (Delta E 76 < 12), filters neutrals and returns the
most frequent cluster. Solves Tailwind/utility-CSS sites where no
semantic --primary variable is exposed.
- Credits: gpt-image-2 metered at 15 credits/image (low quality default).
- Layout: AuthSplitLayout right column is sticky/h-svh so the form
textarea growth no longer stretches the marketing slider.
- i18n cleanup: localized labels follow the no-em-dash convention.
Mentions in post comments
- @mention autocomplete (workspace members, current user excluded) with
marker syntax @[uuid] persisted, display names rendered via CommentBody
chips; live edit replaces markers with names and converts back on save.
- NotifyMentions action with workspace-scoped membership check, dedupes
same user, only newly-added mentions on update.
- Email + in-app via SendNotification job, respecting per-user
notification_preferences.mentioned_in_comment.
- Heartbeat-based presence (Cache, 60s TTL, 30s ping) so online recipients
get only the in-app notification — no email noise.
- Real-time bell on workspace.{id}.user.{id} private channel
(NotificationCreated event), scoped channel name avoids client-side
filtering and lays out a convention for future workspace channels.
- Mailable localized via lang/{en,es,pt-BR}/mail.php; Maizzle source
template for the email is committed and built into resources/views/mail.
AI generation refactor (Action layer + MCP)
- Extracted Actions/Ai/Generate{Image,Video} with QuotaExhaustedException
so agent tools and MCP tools share a single domain entry point.
- Mcp/Tools/Ai/Generate{Image,Video}Tool registered in TryPostServer; both
return MediaResource payloads.
- Orientation::imageApiSize maps non-OpenAI ratios to 1:1/2:3/3:2.
- config/ai.php is now the single source of truth driven by env, removing
the trypost.ai shim. Default text/image providers flipped to OpenAI.
Settings/UX
- /settings/workspace split into shadcn Tabs (Workspace / Brand / Users)
with three components.
- /assets and the in-editor MediaPicker open the ImagePreviewDialog
lightbox on image click while preserving action button behaviour.
- Comments tab landed via ?tab=comments&comment=<id> from notification
click (scroll-to + temporary highlight).
- Mention autocomplete popover flips above when near the viewport bottom.
- Real social platform PNGs replace Tabler brand glyphs in schedule
pills and post list, with hover tooltip carrying display_name + handle.
Bug fixes
- AcceptInvite: controller now passes workspace + role payload that the
Vue page expects; login/register CTAs preselect the invite email.
- WorkspaceInvite mailable: stopped referencing nonexistent
$invite->workspace and $invite->role; column added to the migration,
Invite model casts role to WorkspaceRole, CreateInvite persists it.
- PostCommentCreated: added broadcastAs so .PostCommentCreated actually
matches the Echo listener; payload now includes mentioned_users so
receivers render the chip correctly without a refetch.
- Preview components for X/Pinterest/Threads/Bluesky/LinkedIn/Mastodon/
TikTok/YouTube switched from item.type === 'image' to
!isVideoMedia(item) so media without a persisted type still renders.
- UpdatePostRequest now accepts media.*.{type,mime_type,size,...} so the
posts.media JSON keeps the metadata that the previews need.
- Removed throttle:6,1 from social connect routes (was 429ing legitimate
OAuth retries).
- Used MediaType enum cases instead of literal 'image'/'video' strings
when creating media rows.
Tests
- MentionParser unit tests, NotifyMentions feature tests including
online/offline channel selection and preference gating, MCP AI tool
happy paths, MentionedInComment mailable rendering, AcceptInvite +
search-members + index mentioned_users path. 1229 passing.
When config('trypost.ai.text_provider') points to a provider whose
API key is populated (services.gemini.api_key or
services.openai.api_key), the autofill action pipes the homepage
markdown through a BrandAnalyzer agent with structured output to
produce higher-quality values for:
- brand_description — 2-3 sentences summarizing the company based on
the actual page content, not the raw meta description (which is
often generic SEO boilerplate)
- brand_tone — classified into one of our seven enum values from the
writing style on the page
- content_language — detected from the actual content (more reliable
than <html lang> which is often wrong)
- brand_voice_notes — concrete writing guidelines inferred from the
site's style, written in the detected language
When the LLM provider is NOT configured (open-source self-hosted
deploys without API keys), the action keeps the existing
deterministic meta-tag-only flow — no crash, no noise, no LLM cost.
When the LLM fails mid-request, we log a warning and fall back to
the meta-tag values so the user still gets something useful.
Stack additions:
- league/html-to-markdown ^5.1 converts the main body to clean
markdown for the LLM input (truncated to 4000 chars).
- BrandAnalyzer agent (Agent + HasStructuredOutput) with schema
enums matching our allowed tones and languages.
- resources/views/prompts/brand_analyzer.blade.php holds the
instructions, including explicit enum lists and examples of good
voice_notes.
Frontend: Brand.vue now also fills brand_tone and brand_voice_notes
from the response when present.
Tests (+3): LLM-configured happy path, no-credentials fallback
(asserts BrandAnalyzer is never prompted via preventStrayPrompts),
and LLM-exception fallback (meta tags win, brand_tone stays null).
Users on the Brand step can type their website URL and click
'Preencher' (Portuguese) / 'Autofill' (English). The backend fetches
their homepage, parses standard meta tags, and returns:
- name ← og:site_name | title (suffix stripped at ' | ', ' - ')
- description ← meta[name=description] | og:description
- language ← <html lang> mapped to en / pt-BR / es
- logo ← apple-touch-icon | largest link[rel*=icon] | og:image
The logo is downloaded, validated (mime whitelist, 2MB max), and
attached to the workspace's 'logo' media collection so the avatar
updates immediately.
Zero LLM calls, zero external APIs. Uses symfony/dom-crawler +
symfony/css-selector (newly required) for meta extraction. Everything
else (Http client, workspace media, Intervention) was already in the
project.
Security:
- SSRF guardrail: resolves the host, rejects private / loopback /
link-local ranges, enforces http(s) scheme on both the initial
page fetch and the logo download.
- Rate limited at 10 req/min per user via the throttle middleware
alias on the route.
- Logo content-type must be one of the allowed image mimes; wrong
types are silently dropped so users never see broken images.
UX:
- 'Autofill' button next to the website input, disabled until there
is a URL; shows a spinner while running.
- If a logo was captured, a small preview appears below the input
so users can see what was pulled before saving.
- Success and error paths both surface as vue-sonner toasts, with
translations in en / pt-BR / es.
- Failures leave the form untouched — nothing is destructively
overwritten if parsing gave us nothing.
Tests (16 new): action-level coverage for happy path, title-suffix
fallback, language code normalization across 6 locales, scheme
rejection, private-range SSRF rejection, implicit https prefixing,
empty sites, upstream errors, and wrong-mime logo rejection. Plus
two controller-level tests for the autofill endpoint.
Critical fixes:
- Quota enforcement regressed when the regex flow was replaced with
tools. The LLM was trusted to respect [Session state] quota hints,
which is vulnerable to prompt injection / hallucination. Each tool
now checks Pennant feature limits at the top of handle() and returns
a short quota-exhausted string instead of calling the provider.
Audio shares the video quota (there is no AiAudioLimit by design).
- N+1 on aiMessages: combined the two separate ->get() scans that
produced imagesInThread and videosInThread into a single query.
- postPlatforms lazy load: controller now loadMissing('postPlatforms')
before constructing SocialMediaAssistant, which accesses the relation
in activePlatformRules().
Project conventions:
- \RuntimeException and \Throwable are now imported at the top of the
controller instead of inlined per CLAUDE.md.
- Tool handle() methods use data_get($request, 'prompt') instead of
direct array access, matching the data_get convention for service
classes.
- enrichContent() pluralizes attachment types via Str::plural so
'2 images' reads naturally instead of '2 image'.
Tests:
- Added Storage::fake('public') in tool tests so runs don't pollute
the local disk with fake-generated files.
- Restored quota enforcement coverage (deleted when the flow changed)
as three new tests — one per tool — that fill the monthly bucket
and assert the tool refuses and nothing is generated.
- Removed inline \App\ references in tests in favor of imports.
SocialMediaAssistant::activePlatformRules() reads post_platforms on the
attached Post, maps them to Platform enum values, and resolves rule
classes via Registry::forMany(). The rendered instructions include only
the summaries of platforms actually selected on this post — so when a
user is posting to X only, the agent doesn't waste tokens thinking
about Instagram Reels.
The PLATFORM KNOWLEDGE block in system.blade.php is replaced with
@include('prompts.assistant.platforms') which iterates the passed rules
and renders each summary. When the post has no platforms, the section
renders nothing (a @if guards it).
Each platform has a rule class exposing specs() (char limits, aspect
ratios, media limits, format-specific constraints) and summary() (a
short concise description the agent can render into instructions).
Registered in AppServiceProvider::configurePlatformRules() mapping
Platform enum values to rule classes. Both InstagramFacebook and
LinkedInPage share rules with their non-business siblings.
Contract + Registry follow the lookup-map pattern. Registry is a
static registry seeded at boot — rules are cheap enough to new up
per request.
Tests cover all 12 platforms via Pest dataset and verify the forMany
fan-out, clear/register behavior, and representative summaries.
Each tool implements Laravel\Ai\Contracts\Tool with description(),
handle(Request), and schema(JsonSchema). They share a common pattern:
- Constructor receives Workspace, optional Post, userId, and an optional
AttachmentCollector (resolves from container if not injected).
- handle() does the work (Image::of(), Audio::of(), or calls our custom
VideoGenerationService for Veo), persists media + usage log, pushes
the full attachment shape into the collector, and returns a short
text summary to the LLM.
- schema() exposes typed parameters with enum constraints and doc
strings so the LLM selects valid inputs.
Also fix the Ai::extend closure signature — MultipleInstanceManager
passes (app, config) not just (config) to custom creators.
The agent implements Laravel AI SDK's Agent + Conversational contracts:
- instructions() renders the existing Blade system prompt with workspace
brand context (name, description, website, tone, voice notes, locale)
- messages() reads directly from the AiMessage model scoped to the post,
so our existing conversation persistence stays the single source of
truth — no duplicate SDK-managed storage.
- Assistant messages with attachments are enriched inline with a summary
like [This assistant message attached: 2 image] so the model tracks
progress through carousel-style multi-image generations.
- provider() maps the trypost.ai.text_provider config to Lab::Gemini or
Lab::OpenAI, preserving the existing provider-switching behavior.
