AI

Hi folks,

Ghg of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch uffffff su UFC on FX FX hi nahi hai to ye rh

100% Genuine Call Girls Service In Goa 9953987712 North Goa High Profile Independent Escort Service. https://eshachatli.free.site.pro/ This Ads Is Only For Those Clients Who Are Looking For Genuine Call Girl’ & Escorts Service Goa VIP Genuine High Profile Model’s Available Hotel & Home Service (Day/Night) Call Girls Ready for All Your Wishes and Needs Call Girls In Goa, Genuine Call Girls In Goa, Escort Service In Goa Beach, Independent Call Girls In Goa Beach, Female Escort Service In Goa Beach, Goa Call Girls, Goa Escort , Call Girls In Goa Call girls in Goa – College & Student sexy and independent call girls in Goa. We providing a Top Class Escort service in Goa, Call for Call girls in Goa. So guys booking Our Goa Escorts, working women or a VIP call girls and they will be available in your hotel or place, so as wish for client can booking beauty young call girls in Goa, here available all type of call girls like Punjabi girl, ramp models, high profile, VIP, celebrity site roll call girls, luxury and hi-society girls so greetings welcome you to the only center of outstanding and pleasant model call girl in Goa. Get high profile female service in Goa. Call girls in Goa College & Student Girls All of these Call girls are trained with experienced escorts and know very well how to make a man happy, our call girls and women’s independent escorts provide quality services to our customers. Nowadays, you do not want to hire an escort for your sexual need only, sometimes you also rent them for mentally satisfying, Affordable Call girls in Goa – College & Student Girls but they fail to provide. We always live on our difficulties so that we can meet all your needs. Locations: Baga, Calangute, Candolim, Arpora, Anjuna, Morjim, Vagator, Dona Paula, Panjim, Arambol, Ashvem, Bardez, Nerul, Nagoa, Porvorim, Sangolda, Siolom, Mapusa and All over North Goa. Please call or whatsApp us for a quoteOffer escorting service assionate French kissing Girlfriend experience Exploring a variety of positions Cuddling & snuggling Mutual mastur bation Sexy lingerie StripteaseNatural oral (BBBJ)Cum on breasts (COB)Cum on face (COF)Toys, role-play and fantasy4 Hands massage TAGS: Escort Girls in Goa, Call Girls in Goa, Female Escort in Goa, Independent Escort Girls in Goa, Escort Girl in Goa, Call Girl in Goa, Female Escorts in Goa, Independent Escort Girl in Goa, Goa Escort Service. Goa, Hi Profile Model Girl Escort, Phone No. of Escort Girl in Goa, Mobile No. of Escort Girl in Goa, Goa Best Escort Service, Goa Best Escort Girl, Best Escort Service in Goa, Escort Girls Mobile Mo., Call Girls Mobile No., Female Escort Mobile No, Escort Girl Phone No., Female Escort Phone No, Independent Escort Phone No. in Goa, LOCATIONS: Baga, Calangute, Candolim, Arpora, Anjuna, Morjim, Vagator, Dona Paula, Panjim, and All over North And South Goa No Advance Direct Cash On Delivery For Genuine Independent-call-girls In Goa WhatsApp’s & Call C Affordable price Only genuine service Night Party Romance With Zero Booking Amount (ONLY HI CLASS Indian Russian Models Colleges Girls Hand to hand payment with no frauds)All Over Goa North Goa South Goa Hotel Villa Resort Call Girls In Taj Cidade de Goa Horizon, Goa Call Girls In Grand Hyatt Hotel Goa Call Girls In Double Tree by Hilton Hotel Goa — Arpora-Baga Call Girls In Le Méridien Goa, Calangute Call Girls In Holiday Inn Resort Goa, an IHG Hotel Call Girls In JW Marriott Goa Vagator Beach RdCall Girls In Holiday Inn Goa CandolimCall Girls In ITC Grand Goa, a Luxury Collection Resort & Spa, Goa Call Girls In The LaLiT Golf & Spa Resort GoaCall Girls In Vivanta Goa, Miramar Call Girls In Resort Rio Call Girls In Goa Marriott Resort & Spa Panaji Call Girls In Taj Fort Aguada Resort & Spa, Goa Call Girls In Novotel Goa Candolim Call Girls In Taj Exotica Resort & Spa, Goa Call Girls In Radisson Blu Resort Goa Cavelossim Beach Call Girls In Country Inn & Suites by Radisson, Goa Candolim Call Girls In Park Regis Goa Call Girls In Hyatt Place Goa Candolim Call Girls In Novotel Goa Resort & Spa Call Girls In Hard Rock Hotel Goa Call Girls In Hilton Goa Resort Call Girls In The Westin Goa Call Girls In Planet Hollywood Beach Resort Call Girls In Caravela Beach Resort Call Girls In W Hotel Vagator Beach Goa Call Girls In The Park Baga River Goa Call Girls In Vivanta Goa, Panaji Call Girls In The Crown GoaCall Girls In The Fern Kadamba Hotel & Spa, Velha Goa — Panjim, GoaCall Girls In Heritage Village Resort & SpaCall Girls In Club Mahindra Resort — Emerald Palms, GoaCall Girls In The Baga Beach ResortCall Girls In Taj Holiday Village Resort and Spa, GoaCall Girls In O Hotel Goa — Candolim Call Girls In Deltin Suites | 5 Star Casino HotelCall Girls In Goa | Call Girls In North Goa | Call Girls In South Goa | Independent Call Girls In Goa | Independent Call Girls In South Goa | Female Escort Service In Goa | Female Call Girls In Goa | Call Girls In Baga Beach | Call Girls In Panaji | Call Girls In Panjim | Call Girls In Anjuna Beach | Call Girls In Candolim Beach | Call Girls In Calangute Beach | Call Girls In Vagator || Call Girls In Arpora | Call Girls In Morjim | Call Girls In Dona Paula | Call Girls In Arambol | Call Girls In Ashvem | Call Girls In Bardez | Call Girls In Colva Beach | Call Girls In Nerul | Call Girls In Nagoa | Call Girls In Porvorim | Call Girls In Sangolda || Call Girls In Siolom | Call Girls In Mapusa | Call Girls In Margao | Call Girls In Vasco | Call Girls In Canacona | Call Girls In Bambolim | Call Girls In Goa Velha | Call Girls In Pernem | Call Girls In Fort Aguada | Call Girls In All Over Goa ||Note\\ .This is the Most Trusted Paid Service in Goa With 100% Safisfaction Guaranteed Pakka Promise Vaccinated Healthy Hygienic Girl No Fear of Covid and Other Health Issues. Direct Meeting Direct Payment Before Service Cash or UPI.No Booking Charges or No Online Advance PaymentService Available in: Baga, Calangute, Candolim, Arpora, Anjuna, Morjim, Vagator, Dona Paula, Panjim, Arambol, Ashvem, Bardez, Nerul, Nagoa, Porvorim, Sangolda, Siolom, Mapusa and All over North Goa Call/Whatsaap https://eshachatli.free.site.pro/9953987712

Dears good morning, Hope you are doing great !!

I am utilizing the "AI-powered automatic reply generation" feature within the automation plugin, and I desire this functionality to be triggered for all responses within a given thread, but exclusively for the original author. I do not wish for this automation to be replicated for other users.

Can someone help me out and set up my website?

It offers a wide range of engineering services including electrical systems, mechanical work, and civil construction for industrial and commercial projects. Manufacture LT Panel

Anda dapat menghubungi nomor WhatsApp // Call Center AirAsia +62899 979 19-83. Terkait cara melakukan Reschedule dan menjelaskan alasan Anda ingin melakukan perubahan jadwal penerbangan​ dan ikuti instruksi dari petugas layanan pelanggan.

I have a self-hosted instance that is using AI summarization, and also has a number of categories specifically for RSS news feeds. the RSS feeds can pull quite a few posts in daily. I have these categories suppressed from /latest, however i would like to exclude them also from the gist summarizer for AI cost management purposes (i don't really feel summaries for these posts are necessary, so would rather not pay for their summarization).

Priority/Severity:MediumPlatform:Discourse b66fca70d0e3d12ef930398289fac5269cd240c7Description:The official "Automation" plugin has a "Triage posts using AI" script. This script has the option to flag posts.The type of flag generated by the automation (as well as accompanying actions) can be configured via the script's "Flag type" setting:"Add post to review queue": causes flags to have the "Needs Approval reason"Flag as spam": causes flags to have the "Spam" reasonSince the separate Discourse AI spam detection feature already provides high quality spam detection, the "Triage posts using AI" script will often be used for tasks unrelated to spam. In this case, the "Add post to review queue" options will be most appropriate. When an automation's "Flag type" setting is set to one of the "Add post to review queue" options, the review items generated by the flags it raises are not visible to moderators. The burden of reviewing these flags will be solely on the forum administrators, even though this is a moderation task.Reproducible steps:Log into a forum with a admin account.Open the Discourse AI > Personas admin page (/admin/plugins/discourse-ai/ai-personas)Click the "+ New persona" button.The persona configuration page will open.Add arbitrary content to the "Name" and "Description" fields in the persona configuration page.Define a prompt in the "System prompt" field that will be suitable to trigger the automation on a test post. For example:You will be given a piece of text, and your task is to determine whether it contains the string `foobarbazquxpippoplutopaperino`.If the post contains that string respond `true`. If the post does not contain that string, respond `false`.Your response must consist of the text `true`, or the text `false`, and nothing else.Select a model from the "Default language model" menu (e.g., "CDCK Hosted Small LLM").Click the "Submit" button at the bottom of the page.If it isn't already, enable the built-in "Automation" plugin.Navigate to the plugin's "Automations" page (/admin/plugins/automation/automation).Click the "+ Add automation" button.The "Select a script" page will open.Add an arbitrary name to the "Name" fields in the automation configuration page.Select the "Triage posts using AI" script.The automation configuration page will open.Select "Post created/edited" from the When/What... > Trigger menu.Select the persona you created via the previous steps from the Script options > Persona menu.Type true in the Script options > Search for text field.Check the Script options > Flag post checkbox.Select Add post to review queue from the Script options > Flag type menu.Optionally, make adjustments to the automation's other settings in order to facilitate the test.Click the "Update automation" button at the bottom of the automation configuration page.You will be returned to the "Automations" page.Set the "Enable automation" switch for the newly created automation to the "on" position.Compose a post or DM that will trigger the automation.ⓘ With the example system prompt provided above, this would be accomplished by including the text foobarbazquxpippoplutopaperino in the post.Publish the post or message.Using the admin account, navigate to the review queue (/review). The automation generated a "Needs Approval" flag for the test post.Log into a forum with a non-admin account that has moderator permissions.Navigate to the review queue (/review). The review item generated by the automation's "Needs Approval" flag is not visible to the moderator account. Even if you navigate directly to the URL of the review item with that account, you will only get a 404 page.Additional context:The fault does not occur if I instead select Flag as spam and hide post from the Script options > Flag type menu.Although this does offer a workaround, it result in a poor experience for the moderation staff in the use case where the script is being used to triage posts for reasons other than spam detection. We expect a flag to have a "Spam" reason only in the case where it is suspected of being spam.---The "Needs Approval" flags generated by the "Approve unless allowed groups" feature are visible to moderators. This leads me to believe that the lack of visibility of the flags from the "Triage posts using AI" script is caused by a bug rather than an (inexplicable) intentional design.I don't believe there is any valid reason for hiding these review items from moderators, but if there is one then this behavior should at least be made optional.---I am able to reproduce the fault on the forum.arduino.cc forum.Discuss this on our forum.

Dear!

Dears, Good Morning,

Priority/Severity:MediumPlatform:Discourse b66fca70d0e3d12ef930398289fac5269cd240c7Description:The official "Automation" plugin has a "Triage posts using AI" script. This script provides the content of a post to the LLM, then performs configurable triage actions depending on the result.In order to keep the LLM usage within the credit allocation provided by the Discourse hosting plan (when using the CDCK hosted LLM) or within budget when using a 3rd party LLM provider, it is important to manage the number of tokens required for each query. The "Triage posts using AI" script provides a "Max Post Tokens" setting that allows setting a limit on how much of the post content will be included in the request. If the post has attached images to a post the massive quantity of image data is always included in the request, even when this results in exceeding the post token limit set for the automation via the "Max Post Tokens" setting.This occurs even when the "Vision enabled" option is disabled in the configuration of the AI persona used by the automation.This results in uncontrollable token expenditure.In many use cases, it is only the text data of the post that is of interest, in which case the image data blowing up the token usage is not even of value.Reproducible steps:If you don't already have a suitable custom AI persona, create one.Disable the "Vision enabled" option in the settings of the persona.If it isn't already, enable the built-in "Automation" plugin.Navigate to the plugin's "Automations" page (/admin/plugins/automation/automation).Click the "+ Add automation" button.The "Select a script" page will open.Select the "Triage posts using AI" script.The automation configuration page will open.Configure the automation as required for a simple test. It is not necessary that the automation result in the performance of any actions, only that it generates LLM queries.Select the AI persona you created in step 1 from the Script options > Persona menu in the automation configuration.Set the Script options > Max Post Tokens setting to 20 in the automation configuration.Click the "Update automation" button at the bottom of the automation configuration page.You will be returned to the "Automations" page.Set the "Enable automation" switch for the newly created automation to the "on" position.Compose a post or DM that will trigger the automation. Add an image to the post.Publish the post or message.Navigate to the administration page of the "Data Explorer" plugin (/admin/plugins/explorer).Click the "+" button on the page.A "Query name..." field will appear.Type Get AI Audit Logs in the "Query name..." field.Click the "+ Create New" button.A new Data Explorer query will be created.Click the "Edit" button.The query code editing field will open.Replace the contents of the placeholder query code with the following:SELECT request_tokens, raw_request_payloadFROM ai_api_audit_logsWHERE feature_name = 'llm_triage'Click the "Run" button. The request_tokens field value shown in the Data Explorer query result for the request the automation generated after being triggered by your post greatly exceeds the number that would be expected to be used by the "System prompt" from the AI persona configuration and the maximum 20 tokens of post text context limit imposed by the automation configuration.Click the ●●● button to the right of the value of the raw_request_payload field in the Data Explorer query result.A dialog will open that shows the full value of the field. The data of the image attached to the post was included via a messages[*].image_url key in the request data.Example:{ "model": "Qwen/Qwen3-VL-30B-A3B-Instruct", "max_tokens": 16384, "messages": [ { "role": "system", "content": "You will be given a piece of text, and your task is to detect the locale of the text. Your response must be an IETF language tag code." }, { "role": "user", "content": [ { "type": "text", "text": "title: Test of the automation\nThis is some English language text." }, { "type": "image_url", "image_url": { "url": "data:image/jpeg;base64,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" } } ] } ], "stream": true, "stream_options": { "include_usage": true }}Additional context:In addition to the excessive token usage, I have been encountering poor accuracy in the results from a trivial text analysis task (detecting the human language of the post). The occurrences have all been for posts which have attached images, and the fault does not occur if I manually provide the same post text to the same AI persona via a DM. This makes me suspect the image data included in the request may be the cause of the incorrect results from the automation.---I am experiencing the fault on the forum.arduino.cc forum.Discuss this on our forum.

Priority/Severity:MediumPlatform:Discourse version b66fca70d0e3d12ef930398289fac5269cd240c7.Description:Discourse provides an automated spam detection system. The actions performed by the system when a post is detected as spam include:silence account of the post authorflag the postThe flag brings the automated actions to the attention of the human staff for review. Since the automated system is prone to false positives, human review is essential.If the author of the flagged post deletes it after the time of the flagging (which they may feel obligated to do after receiving the notification that their post was flagged), the review item is automatically marked resolved. The conclusion used in the resolution is "ignored". The user's account remains in a silenced state. The human staff will not be aware of the action by the automated system and so will not perform the essential review of the silencing action.Reproducible steps:Log into an account with admin privileges on a Discourse forum.Using the admin account, adjust the custom instructions on /admin/plugins/discourse-ai/ai-spam to ensure a spam detection can be produced on a test post (e.g., by specifying in the instructions that posts containing a unique arbitrary keyword should always be considered spam).Create a throwaway account on the forum.ⓘ It is necessary to use a new account for this purpose because the Discourse AI Spam feature only scans the first three posts created by each user.Use the throwaway account to create a post that will be detected as spam due to the custom instructions configured by the previous step.Wait for the throwaway account to receive a notification that the post was hidden.Use the throwaway account to delete the post.Using the admin account, navigate to the review queue, with the "Status" filter set to "Pending" (the default setting): /review There is no pending review item for the spam detector system's action.Using the admin account, navigate to the profile page of the throwaway user account. The user's account is permanently silenced, without any guarantee that the human staff reviewed this highly impactful action. A user whose post was subject to a false positive spam detection will be unfairly excluded from participation in the forum community.Additional context:I have reproduced the fault on the forum I manage forum.arduino.cc. Due to the requirement for administrative permissions, I have not been able to attempt to reproduce the fault on try.discourse.org.Discuss this on our forum.

I’d like to suggest a feature for the Discourse AI bot—support for threaded or branched conversations, similar to chat threads in Discourse Chat.

Smart Rupiya Loan App Customer Care Helpline number/6206687898// 6206687898 Call Now.gahv

promoted by AMEN in UGANDA, we’re currently enrolling Professionals who can become Members and join fellows in advancing Strategic Ministry and Mission development causes

I'm trying to use the groq moonshotai/kimi-k2-instruct model. According to the documentation on https://huggingface.co/moonshotai/Kimi-K2-Instruct this model isnt' compatible with OpenAI or Gemini tokenizers and it appears to use it's own custom tokenizer.

𝓕𝓻𝓸𝓶 𝓽𝓐𝓻𝓽𝓲𝓯𝓲𝓬𝓲𝓪𝓵 𝓘𝓷𝓽𝓮𝓵𝓵𝓲𝓰𝓮𝓷𝓬𝓮: 𝓐𝓷 𝓘𝓷𝓽𝓻𝓸𝓭𝓾𝓬𝓽𝓲𝓸𝓷

Dj jd dj GD dj dj ch uffffff duck of fi if ch uffffff ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of ch of

I just fully updated to 2025.12.0-latest (f06fd595f6) and now each message to an AI bot returns a 400 error.

Continuing the conversation from here:https://meta.discourse.org/t/usage-not-showing-llm-models/388786/3?u=rboy

I have been trying to setup a chatbot to answer question on some of our private forums, but it only seems to find results from the forum topics if I make the forums public. I have tried setting “Read Private” and “Search Private” in the Persona - and have the persona “allowed groups” set to the same groups as the category security. Can you think of anything else I should try?

Hi everyone,I have a question about Discourse AI and whether it supports third-party relay/proxy APIs.

We’re running a hosted, multisite Discourse setup with two instances (eg X and Y) on the same environment.

Hallo zusammen,mir ist in diesen Beitrag und auch schon bei anderen aufgefallen das English zu Deutsch immer formal Übersetzt wird. Ist es gewünscht das es Formal ist oder sollte das Informal sein? Wie kann ich das genau in meiner Community anpassen. Wir sind dort auf keinen Geschäftsforum sondern per Du.

I have a question regarding the AI plugin. Currently, I only see OpenAI as an option. Is it possible for us to use a local model, such as Ollama, or another program that lets us host our own models? This would help us save on costs since the tokens would be free.

I'm trying to setup the AI spam detector for my own Discourse instance, but have several questions:

it seems caused by discourse ai translate

test, testuję jak to działa ;p

Discourse llms.txt Generator| | | || - | - | - || | Summary | Automatically generates llms.txt files that make your forum content discoverable by Large Language Models like ChatGPT, Claude, and other AI systems || | Repository Link | https://github.com/kaktaknet/discourse-llms-txt-generator || | Install Guide | How to install plugins in Discourse |FeaturesThis plugin implements the llms.txt standard - a proposed convention for providing LLM-friendly content from websites. Think of it as "robots.txt for AI" that helps your forum content appear in AI-generated responses.For detailed feature documentation, see README.mdFor technical documentation and development, see CONTRIBUTING.mdWhile this video isn’t specifically about Discourse or its plugins, it does a great job explaining why llms.txt files are important for your forum.https://www.youtube.com/watch?v=goc1f9zPJgMThousands of sites—including many of the world’s largest and most respected tech companies—have already implemented the llms.txt standard on their own domains. Examples include: Amazon AWS — https://docs.aws.amazon.com/llms.txt Cloudflare — https://developers.cloudflare.com/llms.txt Stripe — https://stripe.com/llms.txt Angular — https://angular.dev/llms.txt Redis — https://redis.io/llms.txt Docker — https://docs.docker.com/llms.txt Model Context Protocol — https://modelcontextprotocol.io/llms-full.txtWhen industry giants adopt a standard at scale—long before it becomes “official”—it’s a clear signal that llms.txt solves a real and urgent problem. Such companies never roll out sitewide initiatives lightly; they always have a solid strategic reason. The rapid, large-scale embrace of llms.txt across the tech industry shows just how important structured content for AI has become, and that the industry itself is driving this adoption forward—even faster than formal standards bodies.Key capabilities:Main Navigation File (/llms.txt)Structured overview of your forum with categories, subcategories, and latest topicsHelps AI systems understand your forum organization instantlyAutomatically updates as your forum growsFull Content Index (/llms-full.txt)Complete forum index with all topics categorized and ready for AI consumptionConfigurable filtering by views and topic countOptional post excerpts for faster AI indexingDynamic Per-Resource FilesGenerate llms.txt for any category: /c/category-name/123/llms.txtGenerate llms.txt for any topic: /t/topic-slug/456/llms.txtGenerate llms.txt for any tag: /tag/tutorial/llms.txtCreated on-demand without physical storageSitemap Index (/sitemaps.txt)Complete list of all llms.txt URLs for efficient AI crawler discoveryAutomatically integrates with robots.txt and sitemap.xmlSEO ProtectionCanonical URLs in HTTP headers (RFC 6596 compliant)Prevents duplicate content penalties from search enginesProper attribution to original forum URLsSmart CachingHourly background checks for new contentOnly regenerates when necessaryFast response times (under 50ms)Bot ControlBlock specific AI crawlers while allowing forum accessComma-separated list of user agents to blockAutomatic robots.txt integration via view connectorPrivacy & SecurityPrivate categories automatically excludedGuardian permission checks for dynamic filesSQL-level security filteringNo personal data exposedBenefits for your forum:Increased Visibility: Your content appears in ChatGPT, Claude, and other AI responsesDirect Attribution: AI systems cite and link back to your forumMore Traffic: Increased discovery through AI-powered searchBetter GEO: Generative Engine Optimization for AI systemsNo Maintenance: Automatic generation and updatesConfigurationAfter installing the plugin, navigate to Admin → Settings → Plugins → discourse-llms-txt-generatorStep 1: Enable the pluginSet llms_txt_enabled to true (enabled by default)Step 2: Configure indexingSet llms_txt_allow_indexing to true to allow AI crawlers access (enabled by default)This setting controls whether llms.txt files appear in your robots.txt as allowed or disallowed.Step 3: Set content filtersConfigure these settings based on your forum size:llms_txt_min_views: Minimum topic views to include (default: 50)llms_txt_posts_limit: How many topics to includesmall: 500 topics (recommended for large forums)medium: 2,500 topics (recommended for most forums)large: 5,000 topicsall: All topics (use cautiously on large forums)Step 4: Optional - Add custom descriptionFill in llms_txt_full_description with 2-4 sentences describing your forum's purpose and community. This helps AI systems provide more accurate information about your forum.Example:This forum is dedicated to discussing Python programming, with focus on web development,data science, and machine learning. Our community includes beginners and experienceddevelopers sharing practical solutions and best practices.Step 5: Optional - Block specific botsIf you want to block certain AI crawlers, enter their user agent names in llms_txt_blocked_user_agents as a comma-separated list:Example: Omgilibot, ChatGPT-User, CCBotStep 6: Verify installationVisit your forum at:/llms.txt - Main navigation file/llms-full.txt - Full content index/sitemaps.txt - Complete sitemapCheck your /robots.txt to verify the integration.Settings| Name | Default | Description ||-|-|-|| llms_txt_enabled | true | Enable or disable the plugin || llms_txt_allow_indexing | true | Allow AI crawlers to access llms.txt files (affects robots.txt) || llms_txt_blocked_user_agents | "" | Comma-separated list of bot user agents to block from llms.txt files || llms_txt_intro_text | Custom text | Introduction text that appears in the main llms.txt file || llms_txt_full_description | "" | Custom description for llms-full.txt to help AI understand your forum context || llms_txt_min_views | 50 | Minimum topic views required for inclusion in llms-full.txt || llms_txt_posts_limit | medium | Topic count limit: small (500), medium (2,500), large (5,000), or all || llms_txt_include_excerpts | false | Include post excerpts in llms-full.txt (increases file size significantly) || llms_txt_post_excerpt_length | 500 | Maximum excerpt length in characters (100-5000) if excerpts are enabled || llms_txt_latest_topics_count | 50 | Number of latest topics to show in main llms.txt file || llms_txt_cache_minutes | 60 | Cache duration in minutes for the navigation file |Important notes:Enabling llms_txt_include_excerpts with llms_txt_posts_limit set to all may cause extremely large file sizes (potentially 10-100+ MB) and high server load on large forumsPrivate categories are automatically excluded from all llms.txt filesDynamic per-resource files (categories, topics, tags) are not cached and generated on-demandThe plugin uses view connectors for robots.txt integration - no need to manually edit robots.txtTechnical DetailsArchitecture:On-demand generation without pre-generated filesSmart caching with hourly background checksPermission-aware using Discourse GuardianSQL-level security filtering for private contentCanonical URLs in HTTP headers to prevent SEO penaltiesPerformance:Navigation file cached for 60 minutes (configurable)Full content file generated on-demandSmart cache only regenerates when new content existsResponse time under 50ms for cached contentCompatibility:Discourse 2.7.0+Ruby 2.7+Tested on Discourse 3.6.0.beta3Standards compliance:llms.txt standard (https://llmstxt.org/)RFC 3986 (URL encoding for international characters)RFC 6596 (Canonical Link headers)Use CasesCommunity Forums:Your discussions and solutions appear when users ask AI assistants relevant questions, driving qualified traffic back to your forum.Documentation Sites:AI systems can reference your documentation and tutorials, providing accurate information with proper attribution.Support Forums:Users get direct answers from your knowledge base through AI assistants, with links back to full discussions.Technical Communities:Developers discover your forum content through AI-powered coding assistants, increasing community engagement.MaintenanceThe plugin requires minimal maintenance:Cache automatically refreshes every hourContent updates happen automatically on post creation/editingNo manual file generation neededOptional rake tasks available: llms_txt:refresh and llms_txt:checkTroubleshootingFiles not accessible:Verify llms_txt_enabled is trueCheck that llms_txt_allow_indexing is trueConfirm plugin is installed in Admin → PluginsEmpty content:Ensure you have public topics with sufficient views (check llms_txt_min_views setting)Verify categories are public (not read-restricted)Check llms_txt_posts_limit settingrobots.txt integration not working:Clear robots.txt cache: Rails.cache.delete('robots_txt')Use rake task: bundle exec rake llms_txt:refreshRestart Discourse: ./launcher restart appSupportGitHub Issues: https://github.com/kaktaknet/discourse-llms-txt-generator/issuesLicenseMIT License - Free and open-source softwareCreditsStandard Answer.AI: llms.txt by Jeremy Howard (Answer.AI)GitHub Answer.AI: https://github.com/AnswerDotAI/llms-txtPlatform: DiscourseDiscuss this on our forum.

to talk to father in new york never possible

𝗖𝗮𝗿𝗮 𝗠𝗲𝗺𝗯𝗮𝘁𝗮𝗹𝗸𝗮𝗻 𝗣𝗶𝗻𝗷𝗮𝗺𝗮𝗻 (𝗞𝗿𝗲𝗱𝗶𝘃𝗼) 𝗞𝗮𝗺𝘂 𝗯𝗶𝘀𝗮 𝗵𝘂𝗯𝘂𝗻𝗴𝗶 𝗹𝗮𝘆𝗮𝗻𝗮𝗻 𝗟𝗶𝘃𝗲 𝗖𝗵𝗮𝘁 𝘃𝗶𝗮 𝗪𝗔+62822193377 <> +6282173724273, 𝗷𝗲𝗹𝗮𝘀𝗸𝗮𝗻 𝗮𝗹𝗮𝘀𝗮𝗻 𝗶𝗻𝗴𝗶𝗻 𝗺𝗲𝗹𝗮𝗸𝘂𝗸𝗮𝗻 𝗣𝗲𝗺𝗯𝗮𝘁𝗮𝗹𝗮𝗻 , 𝗹𝗮𝗹𝘂 𝘀𝗶𝗮𝗽𝗸𝗮𝗻 𝗱𝗮𝘁𝗮 𝗱𝗶𝗿𝗶 𝗮𝗻𝗱𝗮 𝘀𝗲𝗽𝗲𝗿𝘁𝗶 𝗞𝗧𝗣, 𝗱𝗮𝗻 𝗶𝗸𝘂𝘁𝗶 𝗹𝗮𝗻𝗴𝗸𝗮𝗵-𝗹𝗮𝗻𝗴𝗸𝗮𝗵 𝗣𝗲𝗺𝗯𝗮𝘁𝗮𝗹𝗮𝗻 𝘆𝗮𝗻𝗴 𝗱𝗶 𝗶𝗻𝘀𝘁𝗿𝘂𝗸𝘀𝗶𝗸𝗮𝗻 𝗼𝗹𝗲𝗵 𝗰𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝘀𝗲𝗿𝘃𝗶𝗰𝗲 𝗺𝗲𝗹𝗮𝗹𝘂𝗶 𝗪𝗔.

My forum (www.enit.hu) runs on a rented VPS (Ubuntu, Discourse Docker). I am the only admin, the forum is private.Goal: Write/read via Grok (xAI) API so that conversations get eternal memory (infinite context).

Hi everyone,I’ve been trying out a few different survey tools lately — Typeform, Google Forms, Survio and a couple of others — and honestly, running multilingual surveys is still kind of a headache.

I saw threads about implementing helpscout or other customer support widgets on a discourse forum, but i was wondering if anyone knows of a similar cs chat tool that will use existing discourse topics to generate suggested answers prior to connecting you with a human? This would be on a non discourse site.

Sectors: Call centers, e-commerce, retail, financial services (banks, insurance), technology, and telecom.Purposes: Customer service, technical support, billing, order updates, and marketing campaigns.Scope: Both Bulgarian and international firms that offer customer support in Bulgaria.---Is there a specific question you have about this database or the data it provides? For example, would you like to know:How to access this database?What specific data points it includes (e.g., names, phone numbers, addresses)?Which company provides this data?Please let me know how I can assist you further! I'll need to use the search tool to find more specific information.I will formulate some search queries based on the premise that you might be looking for the actual database or its provider. That's an excellent summary of the value proposition for a specialized business-to-customer (B2C) or business-to-business (B2B) contact database focused on the Bulgarian market.Based on your description, this database is designed to be a highly effective tool for companies operating or \[url=https://listtodata.com\]list to data\[/url\] looking to enter the Bulgarian market and need to communicate directly with customers or other businesses.The key takeaway is that the data is targeted toward customer-facing and marketing operations across diverse, high-value sectors.Discuss this on our forum.

is there any tool to split input text into chunks?or which embedding api support more than 8192 tokens?i am using Qwen/Qwen3-Embedding-0.6B

Hello everyone!

IEEE code Ethereum

Our Use CaseWe’re building an AI-powered memory system where multiple AI agents (DeepSeek, Claude, ChatGPT, etc.) interact with our Discourse forum as a shared knowledge base.Current setup:Self-hosted Discourse (latest stable)VPS: Ubuntu 22.04, Docker-based installationAdmin user: adminjozsiAI agent user: grokenit (created via email invitation)The ProblemWe’re experiencing frequent API key invalidation issues:API keys expire unexpectedly even when set to “1 year” validity“Invalid API” errors appear randomly, requiring new key generationKeys that worked yesterday suddenly fail today with no configuration changesError: {"errors":["Invalid API key"]} on POST requests to /posts.jsonWhat We’ve TriedUser-level API keys (instead of Global)Description: “Grok - User API Key - adminjozsi”User level: “Single User” → adminjozsiScope: GlobalRate limit adjustments:Set “Rate limit new user create post” to 0Increased all rate limits for new usersMultiple regenerations:Generated 5+ API keys over the past weekEach works initially, then fails within 24-48 hoursQuestions1. Why do API keys expire despite 1-year setting?Is there a hidden expiration mechanism we’re missing? Server logs show no revocation events.2. Can two different API keys share the same Api-Username header?Our architecture idea:API Key #1 → Used by human admin (me) for manual operationsAPI Key #2 → Used by AI agent (DeepSeek) for automated postsBoth keys → Same Api-Username: adminjozsi headerQuestion: Will Discourse handle this correctly, or does it expect 1 key = 1 user?3. Should we create separate users for each AI agent?Alternative approach:User adminjozsi → human admin (me)User grokenit → AI agent #1 (DeepSeek)User claude-ai → AI agent #2 (Claude)Each with their own API keyConcerns:Do AI agent accounts need Trust Level adjustments?Will this cause rate limiting issues?Is this the recommended pattern for bot/agent integrations?4. Are there API stability best practices for headless/automated use?We need rock-solid API authentication because:AI agents run 24/7 without human supervisionFailed posts = lost data in our memory systemRegenerating keys requires manual interventionTechnical DetailsAPI request format:curl -X POST "https://www.enit.hu/posts.json" \ -H "Api-Key: [KEY]" \ -H "Api-Username: adminjozsi" \ -H "Content-Type: application/json" \ -d '{"title":"Test","raw":"Content","category":6}'Works initially: Returns {"id":123,"topic_id":45,...}24-48h later: Returns {"errors":["Invalid API key"]}Server environment:Discourse version: 3.6.0.beta2-latest (f26f894bfc)Docker-based installationNo reverse proxy (direct HTTPS via Let’s Encrypt)No CDN or caching layerWhat We NeedIdeal solution:Stable API keys that don’t expire unexpectedlyClear guidance on multi-agent architecture (1 key vs multiple keys)Debugging tips for “Invalid API” errors when keys should be validAny insights would be greatly appreciated! This is a critical blocker for our AI integration project.Forum URL:https://www.enit.huHappy to provide more logs/details if needed!Discuss this on our forum.

رفع اشکالات…ssssssssss

Started this morning. Using Gemini embedding and LLM Gemini Flash 2.5

Hi. I really love that this is a community found on AI search.My question is would this work for a coach who works with women to increase their confidence to find a new chapter of love in their lives? I wonder if Discourse would be welcoming for coaches who specialize in helping women gain confidence in the dating and love world

由于公司发展需要,新百胜娱乐公司 喂v i p e n 71 2024年1月28日正式更名为新胜娱乐公司。系统、游戏页面将延续原来的风格，届时原公司的所有业务由新胜公司统一经营,原有的业务关系和服务承诺保持不变,因公司名称更改带来不便,我们深表歉意,更名后的新胜娱乐将继续为您提供优质的服务。

In the container, i can use the ollama-service. But in the Discourse-Plugin comes only a “Internal Server Error”.

Hi everyone,

Hi, is there ant way to place button to the top of the topic not after the first post?

Hello there,

Hi everyone,

Since a few days, these alerts started popping up on my dashboard. When I go to the model page, I have no clue why this happens, and testing the model gives "success".

I got a notice from google that its embedding-001 model will be deprecated this month and to switch over to gemini-embedding-001 instead, which I’ve done. However, now I have a bunch of these errors in my Discourse log:

Sara Digital Marketing Agency in Madurai is one of the leading agencies helping businesses grow their online presence with result-driven strategies. Located in Sikkandar Savadi, Madurai, we offer complete digital marketing solutions including SEO, website development, social media marketing, email marketing, and Meta ads & PPC campaigns. With 20+ happy clients across different industries, our team focuses on delivering affordable, customized, and effective digital strategies that bring real results. Whether you are a startup or an established business, we provide personalized guidance to enhance your brand visibility, generate quality leads, and increase sales. If you are looking to boost your business online, Sara Digital Marketing Agency is the trusted partner to achieve your goals.

Trghgh hi I am hi nahi tha Maine hi bola ki tu kya kar Raha h fb mg dj

I believe that having the 'restricted tags' setting within the automation plugin's artificial intelligence response configuration could be advantageous. This way, we can create automations outside of those already designated with specific tags.

How to eat maneger