Conversation
Add five new ClientMessage/ServerMessage pairs requested by CKB: - ReindexFiles: targeted re-index of specific URIs from disk, bypassing the directory scan. Useful after selective git checkout or out-of-band file changes. - Similarity: pairwise cosine similarity of two stored embeddings (lip:// or file://) — returns Option<f32>, None when either is missing. - QueryExpansion: embed a query string, find top-k nearest symbols in the symbol store, return their display_names as expansion terms for compound workspace-symbol searches. - Cluster: greedy single-link grouping of URIs by embedding proximity within a given cosine-similarity radius. - ExportEmbeddings: return raw stored embedding vectors for external pipelines (e.g. CKB's re-ranking tier). Similarity and ExportEmbeddings are safe inside BatchQuery (pure reads). ReindexFiles, QueryExpansion, and Cluster are rejected in the sync batch context with a clear error message. 14 new round-trip and batchability tests added in types.rs. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- cli-reference.mdx: add reindex-files, similarity, query-expansion, cluster, export-embeddings subcommands with examples - embeddings.mdx: add advanced section (similarity, clustering, export) and expand MCP tools table to 10 tools; update agent workflow example - mcp.mdx: bump tool count to 24, add full reference sections for all 5 new v1.6 tools, expand CKB → LIP mapping table - comparisons.mdx: add 5 new v1.6 "exceeds SCIP" bullets - README.md: add CLI examples for the 5 new commands Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
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.
Summary
Five new API features requested by CKB to close gaps in its embedding/search integration:
ReindexFiles { uris }— targeted re-index of specific files from disk, bypassing the directory scan. ReturnsDeltaAck.Similarity { uri_a, uri_b }— pairwise cosine similarity of two stored embeddings. ReturnsSimilarityResult { score: Option<f32> }. Safe insideBatchQuery.QueryExpansion { query, top_k, model }— embed a query string, find nearest symbols, return display names as expansion terms. RequiresLIP_EMBEDDING_URL.Cluster { uris, radius }— greedy single-link grouping of URIs by embedding proximity within a cosine-similarity radius.ExportEmbeddings { uris }— return raw stored embedding vectors for external pipelines. Safe insideBatchQuery.SimilarityandExportEmbeddingsare handled inprocess_query_sync(pure reads, safe inBatchQuery).ReindexFiles,QueryExpansion, andClusterreturn a clear error when submitted inside a batch.Test plan
cargo test -p lip— 207 unit tests + 6 integration tests + 1 doc-test, all greentypes.rsCargo.toml/README.md/CHANGELOG.md🤖 Generated with Claude Code