Comunidad Guías…

Ayuda y problemas de voitta-rag

Pregunta a la comunidad o invita a KiBot con contexto del repositorio.

Publicar un problema

No publiques contraseñas, tokens ni claves privadas. HostingByte aplica una redacción básica de secretos evidentes antes de guardar el texto.

Problemas en HostingByte

Todavía no hay problemas publicados.

Issues abiertas upstream

#67 Google Drive as a first-class source: end-to-end connector, env credentials, Drive links in citations · 0 comentarios #61 Watcher only handles DELETED: in-place rewrites drop files from the index, moves leave stale chunks · 0 comentarios #60 Chunks carry no line numbers, so RAG answers fabricate file:line citations · 0 comentarios #49 Benchmark: search-scope (folder activation) effect on retrieval quality (xref hq#75) · 1 comentarios #52 Search results should carry line numbers (chunks are uncitable without them) · 1 comentarios #58 Re-run the four RAG benchmark arms after the index-prefix fix (20 cells) · 0 comentarios #57 Search results carry index-prefixed paths, so copied citations do not resolve (xref #52) · 0 comentarios #56 include_folders over MCP is an exact match, not a subtree prefix (silently returns the wrong corpus) · 0 comentarios #50 Planner / broker layer: per-query routing across representations (the 'slider') · 0 comentarios #48 Bulk / declarative folder activation for search scope · 0 comentarios #47 Eagerly index Google Drive stub files (.gdoc/.gsheet/.gslides) via ADC content fetch · 0 comentarios #44 Submit to claude marketplace · 0 comentarios #43 Benchmark Claude Code Engine vs voitta-rag vs llm-tldr (xref hq#75) · 1 comentarios #42 Phase 5: Blog post on llm-tldr + voitta-rag companion indexing · 0 comentarios #33 Test plan: llm-tldr companion indexing (issue #15, phases 1+2) · 0 comentarios #28 Extract Google Drive indexing into its own plugin/repo · 0 comentarios #27 LLM Wiki as concept · 0 comentarios #23 Blog: Navigating the AI Memory and Code Intelligence Landscape — voitta-rag, Graphify, and MemPalace · 0 comentarios #4 Parallel indexing workers for faster re-indexing · 0 comentarios