[BUG]: [BUG]: Chat with Elara - AI Book Recommendation Engine fails to process semantic mood context #574

Description

Description:
While auditing the Bibliodrift workspace application, I identified a critical functional defect in the AI chatbot module (/chat). The system prompts users to input emotional contexts, but the recommendation handler fails to parse the incoming string parameters or query the dataset, collapsing into an immediate fallback loop.

Steps to Reproduce

Navigate to the /chat workspace of the Bibliodrift interface.

Observe the contextual hint element: Try: "I want something that feels like a letter from a friend".

Type a relevant, descriptive emotional mood query (e.g., "My heart is a little heavy today, I need something that understands grief").

Submit the input query.

Observe that instead of rendering a carousel of matching book item components, the assistant displays a generic error fallback.

Expected Behavior

Expected Behavior:
The conversational handler should intercept the semantic intent of the mood input, cross-reference it with the available book database indexes, and return dynamic book recommendation elements.

Actual Behavior

Proposed Solution:
Audit the backend prompt structure or endpoint routing handler managing the LLM pipeline to ensure sentiment analysis is functioning properly.

Check the backend array logic to guarantee that queries map to fallback genres (e.g., matching "grief" to "Drama/Philosophical Fiction") if a direct keyword match isn't available in the dataset repository.

Assignment:
I have reproduced this core algorithmic failure and would like to fix the recommendation engine parser.

Please assign this task to me under GSSoC 2026.

Screenshots / Logs

OS

No response

Browser

No response

Version

No response