bot.py now calls orchestrate_retrieval() from retrieval.py instead of doing keyword retrieval and vector search separately then concatenating. Claims found by both systems get RRF-boosted to the top. Query decomposition (multi-part → sub-queries) is now active. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| bot.py | ||
| eval_checks.py | ||
| kb_retrieval.py | ||
| market_data.py | ||
| response.py | ||
| retrieval.py | ||
| teleo-telegram.service | ||
| worktree_lock.py | ||
| x_client.py | ||
| x_search.py | ||