Repository navigation
Index local documents and cite the file and passage - #9
Merged
Merged
Conversation
cursor
Bot
force-pushed
the
cursor/knowledge-base-3ce6
branch
from
October 10, 2026 04:08
163e424 to
35d3cfc
Compare
GoDeskio
marked this pull request as ready for review
October 10, 2026 04:49
PDF, DOCX, Markdown, text, and HTML stay in a per-user SQLite index. Local ONNX embeddings come from Hugging Face with checksum checks, and search cites the file and passage. Co-authored-by: Admin <GoDeskio@users.noreply.github.com>
cursor
Bot
force-pushed
the
cursor/knowledge-base-3ce6
branch
from
October 10, 2026 04:54
35d3cfc to
cb7fe29
Compare
This was referenced Oct 10, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Jonathan Ai can answer from the user's own documents without sending them off the machine.
PDF, DOCX, Markdown, plain text, and HTML are indexed in a per-user SQLite file. Search uses local fastembed/ONNX embeddings from Hugging Face after a checksum check, and falls back to BM25 when that model is not installed. Each answer cites the file and the passage.
/kb add,/kb list,/kb remove, and/kb searchare available from the CLI and the desktop, and the settings panel keeps cloud embeddings off unless the user selects OpenAI.Document text stays on this machine unless that cloud embedding provider is explicitly chosen. The optional extra is
pip install "clawd-codex[kb]". The behavior is adapted from Octop's knowledge base and credited in NOTICE.Product tie: Jonathan Ai answers from the user's own documents on this machine, and each answer cites the file and passage.
Stack order, with GitHub base left on main: merge this PR first, then #10, then #11. This branch is the knowledge-base commit on top of main (
870fbd8). #10 adds only the ACP commit on top of this branch. #11 adds only the channels and devices commits on top of #10.