Commit Graph
1450 Commits
Author SHA1 Message Date
Keegan George b6a88ff03e spec 2025-07-11 12:02:50 -07:00
Keegan George f366ded03b DEV: Force default to be set if it was found not to be set! 2025-07-11 08:19:30 -07:00
Keegan George 87c42145d4 DEV: hide hyde model setting 2025-07-10 15:07:12 -07:00
Keegan George c49f1df2ba DEV: migrate hyde model to content creator persona 2025-07-10 15:00:34 -07:00
Keegan George b46940fc66 DEV: hide reset button
It resets it to default before the migration
2025-07-10 14:46:02 -07:00
Keegan George 249aab100c DEV: Use a simple validator
- no need for seeding checks or module dependant checks
2025-07-10 14:44:02 -07:00
Keegan George 5a29074799 DEV: Use default LLM model 2025-07-10 11:54:35 -07:00
Keegan George 29d415755a fix 2025-07-09 15:42:44 -07:00
Keegan George 5a590fddc1 DEV: updates... 2025-07-09 15:36:34 -07:00
Keegan George cad39c1aa4 DEV: client side 2025-07-09 15:29:50 -07:00
Keegan George 2592be65ff FEATURE: configure a default LLM model for all features
This update adds a setting to the configure a default LLM model to be used for all features (unless overridden).
2025-07-09 15:29:24 -07:00
Natalie Tay d54cd1f602 DEV: Normalize locales that are similar (e.g. en and en_GB) so they do not get translated (#1495)
This commit
- normalizes locales like en_GB and variants to en. With this, the feature will not translate en_GB posts to en (or similarly pt_BR to pt_PT)
- consolidates whether the feature is enabled in `DiscourseAi::Translation.enabled?`
- similarly for backfill in  `DiscourseAi::Translation.backfill_enabled?`
  - turns off backfill if `ai_translation_backfill_max_age_days` is 0 to keep true to what it says. Set it to a high number to backfill everything
2025-07-09 22:21:51 +08:00
Keegan George cf7288e1bf DEV: Surface topic_id in errors of summarization backfill (#1493)
This update ensures that the `topic_id` related to the error when summarizing is surfaced in the logs, which should help track down the reason for the errors.
2025-07-09 06:39:36 -07:00
Discourse Translator Bot 027d7f1199 Update translations (#1492) 2025-07-09 15:27:19 +02:00
Keegan George 625442af3c FIX: title suggestions should return 5 unique titles (#1491)
This update fixes a regression from https://github.com/discourse/discourse-ai/pull/1484, which caused AI helper title suggestions to begin suggesting numerous non-unique titles because it was looping through structured responses incorrectly.
2025-07-08 06:30:09 -07:00
Natalie Tay 699ea3f501 DEV: Add backtrace to logs (#1489) 2025-07-08 10:39:43 +08:00
Natalie Tay 56f025cf44 FIX: Localize description excerpts as they have limits (#1490) 2025-07-08 10:36:41 +08:00
Rafael dos Santos Silva 7357280e88 FEATURE: Add old OpenAI tokenizer to embeddings (#1487) 2025-07-07 15:07:27 -03:00
Natalie Tay 6f8960e549 FIX: Pass topic to context (#1488) 2025-07-07 14:59:48 +08:00
Rafael dos Santos Silva 6247906c13 FEATURE: Seamless embedding model upgrades (#1486) 2025-07-04 16:44:03 -03:00
Sam ab5edae121 FIX: make AI helper more robust (#1484)
* FIX: make AI helper more robust

- If JSON is broken for structured output then lean on a more forgiving parser
- Gemini 2.5 flash does not support temp, support opting out
- Evals for assistant were broken, fix interface
- Add some missing LLMs
- Translator was not mapped correctly to the feature - fix that
- Don't mix XML in prompt for translator

* lint

* correct logic

* simplify code

* implement best effort json parsing direct in the structured output object
2025-07-04 14:47:11 +10:00
Joffrey JAFFEUX 0f904977a4 FIX: correctly translate automation name (#1485)
* FIX: correctly translation automation name

The key is persona, not name.

* simplify
2025-07-04 13:01:45 +10:00
Natalie Tay 2b9a4f9232 FIX: Ignore captions and quotes when detecting locale and update prompts (#1483)
A more deterministic way of making sure the LLM detects the correct language (instead of relying on prompt to LLM to ignore it) is to take the cooked and remove unwanted elements.

In this commit 
- we remove quotes, image captions, etc. and only take the remaining text, falling back to the unadulterated cooked
- and update prompts related to detection and translation
- /152465/12
2025-07-03 22:57:48 +08:00
moin-Jana 8b4f401a7b FIX: Typo in custom_tool_exists text (#1480) 2025-07-03 10:17:31 -03:00
Discourse Translator Bot 92e3615378 Update translations (#1479) 2025-07-02 22:36:46 +02:00
Rafael dos Santos SilvaandRoman Rizzi d792919ddf DEV: Move tokenizers to a gem (#1481)
Also renames the Mixtral tokenizer to Mistral.

See gem at github.com/discourse/discourse_ai-tokenizers


Co-authored-by: Roman Rizzi <[email protected]>
2025-07-02 14:43:03 -03:00
Roman Rizzi 75fb37144f FEATURE: Use personas for generating hypothetical posts (#1482)
* FEATURE: Use personas for generating hypothetica posts

* Update prompt
2025-07-02 10:56:38 -03:00
Sam 40fa527633 FIX: cross talk when in ai helper (#1478)
Previous to this change we reused channels for proofreading progress and
ai helper progress

The new changeset ensures each POST to stream progress gets a dedicated
message bus channel

This fixes a class of issues where the wrong information could be displayed
to end users on subsequent proofreading or helper calls

* fix tests

* fix implementation (got to subscribe at 0)
2025-07-01 18:02:16 +10:00
moin-Jana 897f31e564 UX: Fix typo in bot.description text (#1474)
Introducing a typo isn't the right way to bypass the check that blocks the term "private messages".

Nice try, though ;)

I changed it to "personal messages".
2025-07-01 17:24:05 +10:00
Yuriy Kurant 8527279594 dev: removes messages section from sidebar (#1477) 2025-07-01 17:23:11 +10:00
Kris 4ad64ed3b6 DEV: replace sortBy with toSorted (#1476) 2025-06-30 16:41:59 -04:00
Roman Rizzi 5ca7d5f256 FIX: Strip uploads from msg when searching for rag fragments (#1475) 2025-06-30 15:03:17 -03:00
Natalie Tay a94daa14e2 FIX: Return no topics when embeddings is disabled (#1473)
When an invalid model is set for embeddings, topics do not load even if embeddings is disabled.

Error:
## RuntimeError in TopicsController#show
Invalid embeddings selected model

This commit checks for valid settings before attempting to load related topics.
2025-06-30 17:45:04 +08:00
Kris 262bd8b145 UX: add filter to features page, update styles (#1471)
* UX: add filter to features page, update styles

* merge fix

* update toggle spec

* test fix
2025-06-30 09:26:53 +10:00
Roman Rizzi 57b00526f8 FIX: Clarify spam response expectations. (#1470) 2025-06-27 16:59:55 -03:00
Roman Rizzi 8d943fa29d FEATURE: Display spam module on features list. (#1469) 2025-06-27 14:18:01 -03:00
Roman Rizzi b35f9bcc7c FEATURE: Use Persona's when scanning posts for spam (#1465) 2025-06-27 10:35:47 -03:00
Sam cc4e9e030f FIX: normalize keys in structured output (#1468)
* FIX: normalize keys in structured output

Previously we did not validate the hash passed in to structured
outputs which could either be string based or symbol base

Specifically this broke structured outputs for Gemini in some
specific cases.

* comment out flake
2025-06-27 15:42:48 +10:00
Sam 73768ce920 FEATURE: Display bot in feature list (#1466)
- allows features to have multiple llms and multiple personas
- sorts module list
- adds Bot as a first class module
- fixes issue where search module was always configured
- some tests
2025-06-27 12:35:41 +10:00
Rafael dos Santos Silva a40e2d3156 FEATURE: Update OpenAI tokenizer to GPT-4o and later (#1467) 2025-06-26 15:26:09 -03:00
Kris 2fe99a0bec UX: add missing translation for uploads (#1464) 2025-06-25 11:36:00 -04:00
Sam 3e74f09d06 FEATURE: improve custom tool infra (#1463)
- Add support for `chain.streamCustomRaw(test)` that can be used to stream text from a JS tool direct to composer
- Add support for llm params in `llm.generate` which unlocks stuff like structured outputs
- Add discourse.createStagedUser, discourse.createTopic  and discourse.createPost - for content creation
2025-06-25 16:25:44 +10:00
Discourse Translator Bot 3cfc749fad Update translations (#1462) 2025-06-24 16:29:23 +02:00
Jarek RadoszandRoman Rizzi 5735f063a3 FIX: A typo in bot filtration in ai-bot-header-icon (#1455)
* FIX: A typo in bot filtration in ai-bot-header-icon

* FIX: Show header icon when there's only one persona with a default LLM set

---------

Co-authored-by: Roman Rizzi <[email protected]>
2025-06-24 10:51:07 -03:00
Kris 757c93e514 UX: make topic list gists link to the topic (#1459) 2025-06-24 09:14:18 -04:00
Natalie Tay 4c1cd5d819 UX: Align llm button in ai features (#1461) 2025-06-24 17:23:58 +08:00
Sam 471f96f972 FEATURE: allow seeing configured LLM on feature page (#1460)
This is an interim fix so we can at least tell what feature is
being used for what LLM.

It also adds some test coverage to the feature page.
2025-06-24 17:42:47 +10:00
dependabot[bot] 1f851bb2e1 bump rack from 3.1.14 to 3.1.16 (#1408)
---
updated-dependencies:
- dependency-name: rack
  dependency-version: 3.1.16
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-06-24 12:42:22 +10:00
Sam 9f2a4094f5 FEATURE: persona/tool import and export (#1450)
Introduces import/export feature for tools and personas.

Uploads are omitted for now, and will be added in a future PR 

*   **Backend:**
    *   Adds `import` and `export` actions to `Admin::AiPersonasController` and `Admin::AiToolsController`.
    *   Introduces `DiscourseAi::PersonaExporter` and `DiscourseAi::PersonaImporter` services to manage JSON serialization and deserialization.
    *   The export format for a persona embeds its associated custom tools. To ensure portability, `AiTool` references are serialized using their `tool_name` rather than their internal database `id`.
    *   The import logic detects conflicts by name. A `force=true` parameter can be passed to overwrite existing records.

*   **Frontend:**
    *   `AiPersonaListEditor` and `AiToolListEditor` components now include an "Import" button that handles file selection and POSTs the JSON data to the respective `import` endpoint.
    *   `AiPersonaEditorForm` and `AiToolEditorForm` components feature an "Export" button that triggers a download of the serialized record.
    *   Handles import conflicts (HTTP `409` for tools, `422` for personas) by showing a `dialog.confirm` prompt to allow the user to force an overwrite.

*   **Testing:**
    *   Adds comprehensive request specs for the new controller actions (`#import`, `#export`).
    *   Includes unit specs for the `PersonaExporter` and `PersonaImporter` services.
* Persona import and export implemented
2025-06-24 12:41:10 +10:00
Roman Rizzi eea96d6df9 FIX: Include JSON instructions in Helper default personas (#1458) 2025-06-23 11:57:50 -03:00