Update src/llm/hugging_face.py

Update src/config_loader.py

Update src/llm/__init__.py

Rename src/llm/meta.py to src/llm/hugging_face.py

Update src/llm/meta.py

Update src/llm/anthropic.py

Update src/prompts.py

Update src/prompts.py

Update src/prompts.py

Update src/prompts.py

Update src/prompts.py

Update src/prompts.py

Update src/prompts.py

Update src/stata/__init__.py

Update src/stata/__init__.py

Create explain.py

Create stata/__init__.py

Update src/query_engine.py

Update src/prompts.py

Update src/verifier.py

Update src/verifier.py

Update src/prompts.py

Update src/config_loader.py

Folder restructure for HF

Update folder structure for HF deployment

Update support for llama/HF models

Update llama model name

Update api-key access for llama model

Update model support:

feat: switch to stable workhorse models (gpt-4.1, sonnet-4-6, gemini-2.5-flash)

fix: handle GPT-5 restricted parameters (no temperature, developer role)

fix: use max_completion_tokens for GPT-5 (replaces deprecated max_tokens)

feat: upgrade to GPT-5 series, increase default max_tokens to 8192

fix: verification prompt now allows SQL data summarization across rows

fix: Round 17 audit — critical/high bug fixes (5 bugs, cross-checked)

fix: Round 15 — 2 HIGH bugs fixed (sql_count citation audit + reader error handling)

fix: Round 14 — 28 Tier 1 + Tier 2 bugs fixed across 17 source files

fix: pre-release audit round 12 — 33 bugs fixed across all tiers

fix: clear both kb_meta.txt and kb_meta_brief.txt on every re-ingestion

fix: curated model lists, reasoning model token scaling, skip temp files

fix: try alternate TEXT columns before word-level fuzzy matching

fix: prevent over-refusal on SQL results with N/A values

feat: brief KB overview for system prompt + original query cross-check

fix: add word-level fuzzy SQL matching for country name variants

fix: SQL fuzzy matching fallback for country names with alternate forms

feat: LLM-generated welcome summary for knowledge base contents

fix: pre-release audit round 9 — anti-hallucination hardening + error handling

fix: strip trailing semicolons from LLM-generated SQL and add SQL-route fallback

fix: Round 8 audit — 15 bugs fixed across 10 files (1 CRITICAL, 7 HIGH, 3 MEDIUM, 4 LOW)

fix: 10 bugs (4 HIGH + 6 MEDIUM) from comprehensive codebase audit