Go to file
Roman Rizzi ff2e18f9ca
FIX: Structured output discrepancies. (#1340)
This change fixes two bugs and adds a safeguard.

The first issue is that the schema Gemini expected differed from the one sent, resulting in 400 errors when performing completions.

The second issue was that creating a new persona won't define a method
for `response_format`. This has to be explicitly defined when we wrap it inside the Persona class. Also, There was a mismatch between the default value and what we stored in the DB. Some parts of the code expected symbols as keys and others as strings.

Finally, we add a safeguard when, even if asked to, the model refuses to reply with a valid JSON. In this case, we are making a best-effort to recover and stream the raw response.
2025-05-15 11:32:10 -03:00
.github/workflows Initial commit 2023-02-17 11:33:47 -03:00
admin/assets/javascripts/discourse DEV: migrates tools form to form-kit (#1204) 2025-04-22 09:23:25 -07:00
app FIX: Structured output discrepancies. (#1340) 2025-05-15 11:32:10 -03:00
assets FEATURE: forum researcher persona for deep research (#1313) 2025-05-14 12:36:16 +10:00
config FEATURE: allow researcher to also research specific topics (#1339) 2025-05-15 17:48:21 +10:00
db FEATURE: forum researcher persona for deep research (#1313) 2025-05-14 12:36:16 +10:00
discourse_automation FEATURE: allow to send LLM reports to groups (#1246) 2025-04-07 15:31:30 +10:00
evals DEV: Support multiple tests per eval and followups per test (#1199) 2025-03-18 11:42:05 +08:00
lib FIX: Structured output discrepancies. (#1340) 2025-05-15 11:32:10 -03:00
public/ai-share UX: improve artifact styling add direct share link (#930) 2024-11-20 13:13:03 +11:00
spec FIX: Structured output discrepancies. (#1340) 2025-05-15 11:32:10 -03:00
svg-icons REFACTOR: update embeddings to formkit (#1188) 2025-03-13 11:27:38 -04:00
test/javascripts FEATURE: Bot Conversation Homepage (#1273) 2025-04-22 10:22:03 -05:00
tokenizers FEATURE: Gemini Tokenizer (#1088) 2025-01-23 18:20:35 -03:00
.discourse-compatibility DEV: Add compatibility with the Glimmer Post Stream (#1230) 2025-04-29 23:55:54 -03:00
.gitignore FEATURE: allow tools to amend personas (#1250) 2025-04-09 15:48:25 +10:00
.npmrc DEV: Switch to use pnpm (#833) 2024-10-14 13:37:20 +02:00
.prettierignore FEATURE: UI to update ai personas on admin page (#290) 2023-11-21 16:56:43 +11:00
.prettierrc.cjs DEV: Update linting configs (#280) 2023-11-03 11:30:09 +00:00
.rubocop.yml DEV: Expose AI spam scanning metrics (#1077) 2025-01-27 11:57:01 +08:00
.streerc DEV: Update linting configs (#280) 2023-11-03 11:30:09 +00:00
.template-lintrc.cjs DEV: Update linting (#326) 2023-11-29 23:01:48 +01:00
Gemfile DEV: Update linting configs (#280) 2023-11-03 11:30:09 +00:00
Gemfile.lock Build(deps-dev): Bump rack from 3.1.12 to 3.1.14 (#1327) 2025-05-12 17:11:57 -03:00
LICENSE DEV: Update license (#1147) 2025-02-24 11:20:06 +08:00
README.md DEV: Extract configs to a yml file and allow local config (#1142) 2025-02-24 16:22:19 +11:00
about.json DEV: GH CI needs discourse-solved (#1220) 2025-03-26 10:12:55 -03:00
eslint.config.mjs DEV: Update eslint config (#917) 2024-11-19 11:57:40 +01:00
package.json DEV: Update linting (#1194) 2025-03-17 15:14:53 +11:00
plugin.rb DEV: Update gems for Ruby 3.4 compatibility (#1281) 2025-04-24 13:02:51 -03:00
pnpm-lock.yaml DEV: Update linting (#1194) 2025-03-17 15:14:53 +11:00
stylelint.config.mjs DEV: Update linting (#1191) 2025-03-13 13:25:38 +00:00
translator.yml UX: Display the indexing progress for RAG uploads (#557) 2024-04-09 11:03:07 -03:00

README.md

Discourse AI Plugin

Plugin Summary

For more information, please see: https://meta.discourse.org/t/discourse-ai/259214?u=falco

Evals

The directory evals contains AI evals for the Discourse AI plugin. You may create a local config by copying config/eval-llms.yml to config/eval-llms.local.yml and modifying the values.

To run them use:

cd evals ./run --help

Usage: evals/run [options]
    -e, --eval NAME                  Name of the evaluation to run
        --list-models                List models
    -m, --model NAME                 Model to evaluate (will eval all models if not specified)
    -l, --list                       List evals

To run evals you will need to configure API keys in your environment:

OPENAI_API_KEY=your_openai_api_key ANTHROPIC_API_KEY=your_anthropic_api_key GEMINI_API_KEY=your_gemini_api_key