AI search engine and MCP server for bots and agents. One Rust binary indexes code, docs, logs, JSONL/CSV and PDFs for BM25 full-text, vector and hybrid search, code search without grep, agent memory and RAG. Elasticsearch-compatible API; Corpus Hub serves hundreds of reference memory corpora.
Rust agent-memoryai-agentsbm25claude-codeclicode-searchcursordatabaseelasticsearchfull-text-searchhybrid-searchjsonllocal-firstmcpmcp-servermodel-context-protocolragrustsearch-enginevector-search
Merge remote-tracking branch 'origin/main' into release/rc99
Cursor Agent2568d79825a4
diff --cc CHANGELOG.mdindex e9fceb062,3ecd09721..dc455b7fa--- a/CHANGELOG.md+++ b/CHANGELOG.md@@@ -204,17 -95,21 +204,26 @@@ mgtf and mariomarcolongohas doc values, or a non-numeric value against a numeric field, reports`valid: false` (with `error`/`explanations`) instead of claiming a body`_search` rejects is valid. The parse-only answer is kept for a selector- that resolves to no index.+ that resolves to no index+ ([#1284](https://github.com/xerj-org/xerj/issues/1284),+ [#204](https://github.com/xerj-org/xerj/issues/204), PR+ [#2263](https://github.com/xerj-org/xerj/pull/2263), Tong-bit-art).+ - On Windows, the process memory cap is limited by commit space (the commit+ limit and what this process can still commit), not only by physical RAM. A+ default `xerj autoindex` on a machine with no pagefile and little free+ commit rejects writes with HTTP 429 naming that limit and how to enlarge+ the pagefile, instead of aborting with `memory allocation of N bytes+ failed` (#1122). The AUTO tier is still chosen from physical RAM, then+ lowered when commit is the tighter ceiling. Linux and macOS budgets are+ unchanged.+- A `semantic` query on a `semantic_text` field returns `highlight` fragments- when the request asks for them (#1947). The fragments are the field's- best-scoring chunks for the query embedding, in score order, limited by- `number_of_fragments` (default 5; `0` returns the whole field). `order:+ when the request asks for them+ ([#1947](https://github.com/xerj-org/xerj/issues/1947), PR+ [#2227](https://github.com/xerj-org/xerj/pull/2227)). The fragments are the+ field's best-scoring chunks for the query embedding, in score order, limited+ by `number_of_fragments` (default 5; `0` returns the whole field). `order:none` emits that same selection in document order. `"type": "semantic"`selects this highlighter, and it is the default for a `semantic_text`field on a semantic query. Chunks are not wrapped in `<em>` tags. A