| docs: add rag setup and eval fixtures | 2 个月前 |
| Add admin RAG Search Lab | 17 天前 |
| Add admin RAG Search Lab | 17 天前 |
| feat: add daemon cache and RAG maintenance lifecycle | 1 个月前 |
| Batch Ollama embeddings and benchmark throughput | 1 个月前 |
| Batch Ollama embeddings and benchmark throughput | 1 个月前 |
| Make source search hybrid by default | 18 天前 |
| fix: harden rag indexing from e2e | 2 个月前 |
| Add admin RAG Search Lab | 17 天前 |
| Add admin RAG Search Lab | 17 天前 |
| Make RAG setup model pulls observable and timeout-safe Closes #106 ## Summary - give confirmed model pulls a long-running timeout instead of the 30-second provider probe timeout - emit an immediate text-mode handoff before a model download begins - re-check provider model state after a pull transport error before reporting failure - print repository-scoped next commands after successful setup - keep JSON setup output a single machine-readable document ## Verification - go test ./... - git diff --check - live dogfood: Ollama model is present, rag setup reports ready, embedding smoke is ok, and hybrid RAG search succeeds | 17 天前 |
| Make RAG setup model pulls observable and timeout-safe Closes #106 ## Summary - give confirmed model pulls a long-running timeout instead of the 30-second provider probe timeout - emit an immediate text-mode handoff before a model download begins - re-check provider model state after a pull transport error before reporting failure - print repository-scoped next commands after successful setup - keep JSON setup output a single machine-readable document ## Verification - go test ./... - git diff --check - live dogfood: Ollama model is present, rag setup reports ready, embedding smoke is ok, and hybrid RAG search succeeds | 17 天前 |
| Make source search hybrid by default | 18 天前 |
| Make source search hybrid by default | 18 天前 |
| Apply hybrid filters before semantic ranking | 18 天前 |
| fix: harden rag indexing from e2e | 2 个月前 |