### Why
This pull request fundamentally restructures how AI bots create and update web artifacts to address critical limitations in the previous approach:
1. **Improved Artifact Context for LLMs**: Previously, artifact creation and update tools included the *entire* artifact source code directly in the tool arguments. This overloaded the Language Model (LLM) with raw code, making it difficult for the LLM to maintain a clear understanding of the artifact's current state when applying changes. The LLM would struggle to differentiate between the base artifact and the requested modifications, leading to confusion and less effective updates.
2. **Reduced Token Usage and History Bloat**: Including the full artifact source code in every tool interaction was extremely token-inefficient. As conversations progressed, this redundant code in the history consumed a significant number of tokens unnecessarily. This not only increased costs but also diluted the context for the LLM with less relevant historical information.
3. **Enabling Updates for Large Artifacts**: The lack of a practical diff or targeted update mechanism made it nearly impossible to efficiently update larger web artifacts. Sending the entire source code for every minor change was both computationally expensive and prone to errors, effectively blocking the use of AI bots for meaningful modifications of complex artifacts.
**This pull request addresses these core issues by**:
* Introducing methods for the AI bot to explicitly *read* and understand the current state of an artifact.
* Implementing efficient update strategies that send *targeted* changes rather than the entire artifact source code.
* Providing options to control the level of artifact context included in LLM prompts, optimizing token usage.
### What
The main changes implemented in this PR to resolve the above issues are:
1. **`Read Artifact` Tool for Contextual Awareness**:
- A new `read_artifact` tool is introduced, enabling AI bots to fetch and process the current content of a web artifact from a given URL (local or external).
- This provides the LLM with a clear and up-to-date representation of the artifact's HTML, CSS, and JavaScript, improving its understanding of the base to be modified.
- By cloning local artifacts, it allows the bot to work with a fresh copy, further enhancing context and control.
2. **Refactored `Update Artifact` Tool with Efficient Strategies**:
- The `update_artifact` tool is redesigned to employ more efficient update strategies, minimizing token usage and improving update precision:
- **`diff` strategy**: Utilizes a search-and-replace diff algorithm to apply only the necessary, targeted changes to the artifact's code. This significantly reduces the amount of code sent to the LLM and focuses its attention on the specific modifications.
- **`full` strategy**: Provides the option to replace the entire content sections (HTML, CSS, JavaScript) when a complete rewrite is required.
- Tool options enhance the control over the update process:
- `editor_llm`: Allows selection of a specific LLM for artifact updates, potentially optimizing for code editing tasks.
- `update_algorithm`: Enables choosing between `diff` and `full` update strategies based on the nature of the required changes.
- `do_not_echo_artifact`: Defaults to true, and by *not* echoing the artifact in prompts, it further reduces token consumption in scenarios where the LLM might not need the full artifact context for every update step (though effectiveness might be slightly reduced in certain update scenarios).
3. **System and General Persona Tool Option Visibility and Customization**:
- Tool options, including those for system personas, are made visible and editable in the admin UI. This allows administrators to fine-tune the behavior of all personas and their tools, including setting specific LLMs or update algorithms. This was previously limited or hidden for system personas.
4. **Centralized and Improved Content Security Policy (CSP) Management**:
- The CSP for AI artifacts is consolidated and made more maintainable through the `ALLOWED_CDN_SOURCES` constant. This improves code organization and future updates to the allowed CDN list, while maintaining the existing security posture.
5. **Codebase Improvements**:
- Refactoring of diff utilities, introduction of strategy classes, enhanced error handling, new locales, and comprehensive testing all contribute to a more robust, efficient, and maintainable artifact management system.
By addressing the issues of LLM context confusion, token inefficiency, and the limitations of updating large artifacts, this pull request significantly improves the practicality and effectiveness of AI bots in managing web artifacts within Discourse.
We have a flag to signal we are shortening the embeddings of a model.
Only used in Open AI's text-embedding-3-*, but we plan to use it for other services.
* Use AR model for embeddings features
* endpoints
* Embeddings CRUD UI
* Add presets. Hide a couple more settings
* system specs
* Seed embedding definition from old settings
* Generate search bit index on the fly. cleanup orphaned data
* support for seeded models
* Fix run test for new embedding
* fix selected model not set correctly
This adds registration and last known IP information and email to scanning context.
This provides another hint for spam scanner about possible malicious users.
For example registered in India, replying from Australia or
email is clearly a throwaway email address.
To quickly select backfill candidates without comparing SHAs, we compare the last summarized post to the topic's highest_post_number. However, hiding or deleting a post and adding a small action will update this column, causing the job to stall and re-generate the same summary repeatedly until someone posts a regular reply. On top of this, this is not always true for topics with `best_replies`, as this last reply isn't necessarily included.
Since this is not evident at first glance and each summarization strategy picks its targets differently, I'm opting to simplify the backfill logic and how we track potential candidates.
The first step is dropping `content_range`, which serves no purpose and it's there because summary caching was supposed to work differently at the beginning. So instead, I'm replacing it with a column called `highest_target_number`, which tracks `highest_post_number` for topics and could track other things like channel's `message_count` in the future.
Now that we have this column when selecting every potential backfill candidate, we'll check if the summary is truly outdated by comparing the SHAs, and if it's not, we just update the column and move on
When enabling spam scanner it there may be old unscanned posts
this can create a risky situation where spam scanner operates
on legit posts during false positives
To keep this a lot safer we no longer try to hide old stuff by
the spammers.
Disabling streaming is required for models such o1 that do not have streaming
enabled yet
It is good to carry this feature around in case various apis decide not to support streaming endpoints and Discourse AI can continue to work just as it did before.
Also: fixes issue where sharing artifacts would miss viewport leading to tiny artifacts on mobile
* FEATURE: smart date support for AI helper
This feature allows conversion of human typed in dates and times
to smart "Discourse" timezone friendly dates.
* fix specs and lint
* lint
* address feedback
* add specs
In a previous refactor, we moved the responsibility of querying and storing embeddings into the `Schema` class. Now, it's time for embedding generation.
The motivation behind these changes is to isolate vector characteristics in simple objects to later replace them with a DB-backed version, similar to what we did with LLM configs.
* FIX: Make sure gists are atleast five minutes old before updating them
* Update app/jobs/regular/fast_track_topic_gist.rb
Co-authored-by: Keegan George <kgeorge13@gmail.com>
---------
Co-authored-by: Keegan George <kgeorge13@gmail.com>
* REFACTOR: A Simpler way of interacting with embeddings' tables.
This change adds a new abstraction called `Schema`, which acts as a repository that supports the same DB features `VectorRepresentation::Base` has, with the exception that removes the need to have duplicated methods per embeddings table.
It is also a bit more flexible when performing a similarity search because you can pass it a block that gives you access to the builder, allowing you to add multiple joins/where conditions.
This introduces a comprehensive spam detection system that uses LLM models
to automatically identify and flag potential spam posts. The system is
designed to be both powerful and configurable while preventing false positives.
Key Features:
* Automatically scans first 3 posts from new users (TL0/TL1)
* Creates dedicated AI flagging user to distinguish from system flags
* Tracks false positives/negatives for quality monitoring
* Supports custom instructions to fine-tune detection
* Includes test interface for trying detection on any post
Technical Implementation:
* New database tables:
- ai_spam_logs: Stores scan history and results
- ai_moderation_settings: Stores LLM config and custom instructions
* Rate limiting and safeguards:
- Minimum 10-minute delay between rescans
- Only scans significant edits (>10 char difference)
- Maximum 3 scans per post
- 24-hour maximum age for scannable posts
* Admin UI features:
- Real-time testing capabilities
- 7-day statistics dashboard
- Configurable LLM model selection
- Custom instruction support
Security and Performance:
* Respects trust levels - only scans TL0/TL1 users
* Skips private messages entirely
* Stops scanning users after 3 successful public posts
* Includes comprehensive test coverage
* Maintains audit log of all scan attempts
---------
Co-authored-by: Keegan George <kgeorge13@gmail.com>
Co-authored-by: Martin Brennan <martin@discourse.org>
Instead of a stacked chart showing a separate series for positive and negative, this PR introduces a simplification to the overall sentiment dashboard. It comprises the sentiment into a single series of the difference between `positive - negative` instead. This should allow for the data to be more easy to scan and look for trends
Instead of a stacked chart showing a separate series for positive and negative, this PR introduces a simplification to the overall sentiment dashboard. It comprises the sentiment into a single series of the difference between `positive - negative` instead. This should allow for the data to be more easy to scan and look for trends.
* FEATURE: first class support for OpenRouter
This new implementation supports picking quantization and provider pref
Also:
- Improve logging for summary generation
- Improve error message when contacting LLMs fails
* Better support for full screen artifacts on iPad
Support back button to close full screen
Refactor dialect selection and add Nova API support
Change dialect selection to use llm_model object instead of just provider name
Add support for Amazon Bedrock's Nova API with native tools
Implement Nova-specific message processing and formatting
Update specs for Nova and AWS Bedrock endpoints
Enhance AWS Bedrock support to handle Nova models
Fix Gemini beta API detection logic
* FEATURE: Backfill posts sentiment.
It adds a scheduled job to backfill posts' sentiment, similar to our existing rake task, but with two settings to control the batch size and posts' max-age.
* Make sure model_name order is consistent.
Add support for versioned artifacts with improved diff handling
* Add versioned artifacts support allowing artifacts to be updated and tracked
- New `ai_artifact_versions` table to store version history
- Support for updating artifacts through a new `UpdateArtifact` tool
- Add version-aware artifact rendering in posts
- Include change descriptions for version tracking
* Enhance artifact rendering and security
- Add support for module-type scripts and external JS dependencies
- Expand CSP to allow trusted CDN sources (unpkg, cdnjs, jsdelivr, googleapis)
- Improve JavaScript handling in artifacts
* Implement robust diff handling system (this is dormant but ready to use once LLMs catch up)
- Add new DiffUtils module for applying changes to artifacts
- Support for unified diff format with multiple hunks
- Intelligent handling of whitespace and line endings
- Comprehensive error handling for diff operations
* Update routes and UI components
- Add versioned artifact routes
- Update markdown processing for versioned artifacts
Also
- Tweaks summary prompt
- Improves upload support in custom tool to also provide urls
- Added a new admin interface to track AI usage metrics, including tokens, features, and models.
- Introduced a new route `/admin/plugins/discourse-ai/ai-usage` and supporting API endpoint in `AiUsageController`.
- Implemented `AiUsageSerializer` for structuring AI usage data.
- Integrated CSS stylings for charts and tables under `stylesheets/modules/llms/common/usage.scss`.
- Enhanced backend with `AiApiAuditLog` model changes: added `cached_tokens` column (implemented with OpenAI for now) with relevant DB migration and indexing.
- Created `Report` module for efficient aggregation and filtering of AI usage metrics.
- Updated AI Bot title generation logic to log correctly to user vs bot
- Extended test coverage for the new tracking features, ensuring data consistency and access controls.
This change adds a simpler class for sentiment classification, replacing the soon-to-be removed `Classificator` hierarchy. Additionally, it adds a method for classifying concurrently, speeding up the backfill rake task.
This PR updates the logic for the location map so it permits only the desired prompts through to the composer/post menu. Anything else won't be shown by default.
This PR also adds relevant tests to prevent regression.
We are adding a new method for generating and storing embeddings in bulk, which relies on `Concurrent::Promises::Future`. Generating an embedding consists of three steps:
Prepare text
HTTP call to retrieve the vector
Save to DB.
Each one is independently executed on whatever thread the pool gives us.
We are bringing a custom thread pool instead of the global executor since we want control over how many threads we spawn to limit concurrency. We also avoid firing thousands of HTTP requests when working with large batches.
This spec fails inconsistently with:
-fragment-n14
+You are a helpful Discourse assistant.
+You _understand_ and **generate** Discourse Markdown.
+You live in a Discourse Forum Message.
+
+You live in the forum with the URL: http://test.localhost
+The title of your site: test site title
+The description is: test site description
+The participants in this conversation are: joe, jane
+The date now is: 2024-11-25 20:23:02 UTC, much has changed since you were trained.
+
+You were trained on OLD data, lean on search to get up to date information about this forum
+When searching try to SIMPLIFY search terms
+Discourse search joins all terms with AND. Reduce and simplify terms to find more results.<guidance>
+The following texts will give you additional guidance for your response.
+We included them because we believe they are relevant to this conversation topic.
+
+Texts:
+
+fragment-n10
+fragment-n9
+fragment-n8
+fragment-n7
+fragment-n6
+fragment-n5
+fragment-n4
+fragment-n3
+fragment-n2
+fragment-n1
+</guidance>
* FEATURE: allow mentioning an LLM mid conversation to switch
This is a edgecase feature that allow you to start a conversation
in a PM with LLM1 and then use LLM2 to evaluation or continue
the conversation
* FEATURE: allow auto silencing of spam accounts
New rule can also allow for silencing an account automatically
This can prevent spammers from creating additional posts.
The `topic_query_create_list_topics` modifier we append was always meant to avoid an N+1 situation when serializing gists. However, I tried to be too smart and only preload these, which resulted in some topics with *only* regular summaries getting removed from the list. This issue became apparent now we are adding gists to other lists besides hot.
Let's simplify the preloading, which still solves the N+1 issue, and let the serializer get the needed summary.
1. Keep source in a "details" block after rendered so it does
not overwhelm users
2. Ensure artifacts are never indexed by robots
3. Cache break our CSS that changed recently
We use `includes` instead of `joins` because we want to eager-load summaries, avoiding an extra query when summarizing. However, Rails will complain unless you explicitly inform them you plan to use that inside a `WHERE` clause.
This is a significant PR that introduces AI Artifacts functionality to the discourse-ai plugin along with several other improvements. Here are the key changes:
1. AI Artifacts System:
- Adds a new `AiArtifact` model and database migration
- Allows creation of web artifacts with HTML, CSS, and JavaScript content
- Introduces security settings (`strict`, `lax`, `disabled`) for controlling artifact execution
- Implements artifact rendering in iframes with sandbox protection
- New `CreateArtifact` tool for AI to generate interactive content
2. Tool System Improvements:
- Adds support for partial tool calls, allowing incremental updates during generation
- Better handling of tool call states and progress tracking
- Improved XML tool processing with CDATA support
- Fixes for tool parameter handling and duplicate invocations
3. LLM Provider Updates:
- Updates for Anthropic Claude models with correct token limits
- Adds support for native/XML tool modes in Gemini integration
- Adds new model configurations including Llama 3.1 models
- Improvements to streaming response handling
4. UI Enhancements:
- New artifact viewer component with expand/collapse functionality
- Security controls for artifact execution (click-to-run in strict mode)
- Improved dialog and response handling
- Better error management for tool execution
5. Security Improvements:
- Sandbox controls for artifact execution
- Public/private artifact sharing controls
- Security settings to control artifact behavior
- CSP and frame-options handling for artifacts
6. Technical Improvements:
- Better post streaming implementation
- Improved error handling in completions
- Better memory management for partial tool calls
- Enhanced testing coverage
7. Configuration:
- New site settings for artifact security
- Extended LLM model configurations
- Additional tool configuration options
This PR significantly enhances the plugin's capabilities for generating and displaying interactive content while maintaining security and providing flexible configuration options for administrators.
Implement streaming tool call implementation for Anthropic and Open AI.
When calling:
llm.generate(..., partial_tool_calls: true) do ...
Partials may contain ToolCall instances with partial: true, These tool calls are partially populated with json partially parsed.
So for example when performing a search you may get:
ToolCall(..., {search: "hello" })
ToolCall(..., {search: "hello world" })
The library used to parse json is:
https://github.com/dgraham/json-stream
We use a fork cause we need access to the internal buffer.
This prepares internals to perform partial tool calls, but does not implement it yet.
This re-implements tool support in DiscourseAi::Completions::Llm #generate
Previously tool support was always returned via XML and it would be the responsibility of the caller to parse XML
New implementation has the endpoints return ToolCall objects.
Additionally this simplifies the Llm endpoint interface and gives it more clarity. Llms must implement
decode, decode_chunk (for streaming)
It is the implementers responsibility to figure out how to decode chunks, base no longer implements. To make this easy we ship a flexible json decoder which is easy to wire up.
Also (new)
Better debugging for PMs, we now have a next / previous button to see all the Llm messages associated with a PM
Token accounting is fixed for vllm (we were not correctly counting tokens)
This change introduces a job to summarize topics and cache the results automatically. We provide a setting to control how many topics we'll backfill per hour and what the topic's minimum word count is to qualify.
We'll prioritize topics without summary over outdated ones.
Fixes encoding of params on LLM function calls.
Previously we would improperly return results if a function parameter returned an HTML tag.
Additionally adds some missing HTTP verbs to tool calls.
* FIX/REFACTOR: FoldContent revamp
We hit a snag with our hot topic gist strategy: the regex we used to split the content didn't work, so we cannot send the original post separately. This was important for letting the model focus on what's new in the topic.
The algorithm doesn’t give us full control over how prompts are written, and figuring out how to format the content isn't straightforward. This means we're having to use more complicated workarounds, like regex.
To tackle this, I'm suggesting we simplify the approach a bit. Let's focus on summarizing as much as we can upfront, then gradually add new content until there's nothing left to summarize.
Also, the "extend" part is mostly for models with small context windows, which shouldn't pose a problem 99% of the time with the content volume we're dealing with.
* Fix fold docs
* Use #shift instead of #pop to get the first elem, not the last
* FIX: Llm selector / forced tools / search tool
This fixes a few issues:
1. When search was not finding any semantic results we would break the tool
2. Gemin / Anthropic models did not implement forced tools previously despite it being an API option
3. Mechanics around displaying llm selector were not right. If you disabled LLM selector server side persona PM did not work correctly.
4. Disabling native tools for anthropic model moved out of a site setting. This deliberately does not migrate cause this feature is really rare to need now, people who had it set probably did not need it.
5. Updates anthropic model names to latest release
* linting
* fix a couple of tests I missed
* clean up conditional
A new feature_context json column was added to ai_api_audit_logs
This allows us to store rich json like context on any LLM request
made.
This new field now stores automation id and name.
Additionally allows llm_triage to specify maximum number of tokens
This means that you can limit the cost of llm triage by scanning only
first N tokens of a post.
* Display gists in the hot topics list
* Adjust hot topics gist strategy and add a job to generate gists
* Replace setting with a configurable batch size
* Avoid loading summaries for other topic lists
* Tweak gist prompt to focus on latest posts in the context of the OP
* Remove serializer hack and rely on core change from discourse/discourse#29291
* Update lib/summarization/strategies/hot_topic_gists.rb
Co-authored-by: Rafael dos Santos Silva <xfalcox@gmail.com>
---------
Co-authored-by: Rafael dos Santos Silva <xfalcox@gmail.com>
Splits persona permissions so you can allow a persona on:
- chat dms
- personal messages
- topic mentions
- chat channels
(any combination is allowed)
Previously we did not have this flexibility.
Additionally, adds the ability to "tether" a language model to a persona so it will always be used by the persona. This allows people to use a cheaper language model for one group of people and more expensive one for other people
This introduces another configuration that allows operators to
limit the amount of interactions with forced tool usage.
Forced tools are very handy in initial llm interactions, but as
conversation progresses they can hinder by slowing down stuff
and adding confusion.
This adds chain halting (ability to terminate llm chain in a tool)
and the ability to create uploads in a tool
Together this lets us integrate custom image generators into a
custom tool.
* FEATURE: allows forced LLM tool use
Sometimes we need to force LLMs to use tools, for example in RAG
like use cases we may want to force an unconditional search.
The new framework allows you backend to force tool usage.
Front end commit to follow
* UI for forcing tools now works, but it does not react right
* fix bugs
* fix tests, this is now ready for review
Previous to this change we could flag, but there was no way
to hide content and treat the flag as spam.
We had the option to hide topics, but this is not desirable for
a spam reply.
New option allows triage to hide a post if it is a reply, if the
post happens to be the first post on the topic, the topic will
be hidden.
This allows our users to add the Ollama provider and use it to serve our AI bot (completion/dialect).
In this PR, we introduce:
DiscourseAi::Completions::Dialects::Ollama which would help us translate by utilizing Completions::Endpoint::Ollama
Correct extract_completion_from and partials_from in Endpoints::Ollama
Also
Add tests for Endpoints::Ollama
Introduce ollama_model fabricator
Polymorphic RAG means that we will be able to access RAG fragments both from AiPersona and AiCustomTool
In turn this gives us support for richer RAG implementations.
Previously we waited 1 minute before automatically titling PMs
The new change introduces adding a title immediately after the the
llm replies
Prompt was also modified to include the LLM reply in title suggestion.
This helps situation like:
user: tell me a joke
llm: a very funy joke about horses
Then the title would be "A Funny Horse Joke"
Specs already covered some auto title logic, amended to also
catch the new message bus message we have been sending.
This improves the site setting search so it performs a somewhat
fuzzy match.
Previously it did not handle seperators such as "space" and a
term such as "min_post_length" would not find "min_first_post_length"
A more liberal search algorithm makes it easier to the AI to
navigate settings.
* Minor fix, {{and parameter.enum parameter.enum.length}} is non
obviously broken.
If parameter.enum is a tracked array it will return the object
cause embers and helper implementation.
This corrects an issue where enum keeps on selecting itself by
mistake.
* FEATURE: LLM Triage support for systemless models.
This change adds support for OSS models without support for system messages. LlmTriage's system message field is no longer mandatory. We now send the post contents in a separate user message.
* Models using Ollama can also disable system prompts
New `ai_pm_summarization_allowed_groups` can be used to allow
visibility of the summarization feature on PMs.
This can be useful on forums where a lot of communication happens
inside PMs.
Sam
Roman Rizzi
Natalie Tay
Rafael dos Santos Silva
Keegan George
Joffrey JAFFEUX
Loïc Guitaut
Hoa Nguyen
Mark VanLandingham