xerj-org/xerj
Local-first search engine and MCP server for AI 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. observed · 2026-10-20
Health v2 · maintenance only
77/100
- Activity 100
- Release rhythm 88
- Longevity 8
Flags: young
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-10-21. Adoption (stars, forks) is never an input.
- gap_med: 0
- age_days: 112
- days_rel: 1
- days_push: 1
- n_releases_24m: 96
Adoption not part of the score
10576 stars · 592 forks observed · 2026-10-20
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded
XERJ is a Rust-based, single-binary search engine built for AI agents, offering BM25, kNN vector, and hybrid search with Elasticsearch-compatible APIs. It indexes code, docs, logs, and PDFs with one zero-config command, enabling token-efficient retrieval, RAG, agent memory, and security analytics without a JVM or separate vector database.
Use cases
- reduce coding agent token usage by retrieving exact code instead of grep
- index a mixed-format folder of logs, PDFs, and CSVs with one command
- give AI agents long-term memory with per-agent isolation
- build semantic search and RAG without a separate vector database
- run SIEM and log analytics on a single node
- drop in an Elasticsearch replacement for existing clients
When to choose
- you want an Elasticsearch-compatible search engine that runs as a single static binary without a JVM
- your AI coding agent wastes tokens reading files via grep and sed
- you need combined BM25, kNN, and hybrid search plus agent memory in one datastore
- you want zero-config indexing of heterogeneous local files
When to avoid
- you rely on whole-index scroll reads or parent/child joins, which are unsupported or capped
- you need a fully managed multi-region cluster today (HA and multi-region tiers are commercial or upcoming)
- you require 100% Elasticsearch API coverage rather than the documented subset
Facets
service · maturity active
search-engine vector-database rag agent-framework mcp logging analytics databases artificial-intelligence developer-tools self-hosted analytics windows rust cli self-hosted elasticsearch-compatible bm25 hybrid-search semantic-search knn agent-memory autoindex token-efficiency single-binary siem log-analytics search ai-agents retrieval-augmented-generation linux macos docker
10 sources
- readme: https://github.com/xerj-org/xerj · fetched 2026-10-20 · 808ae079e78b
- homepage: https://xerj.org · fetched 2026-08-29 · eea2cc69a475
- site_page: https://xerj.org/docs · fetched 2026-08-29 · 9c86035caffa
- site_page: https://xerj.org/docs/recipes · fetched 2026-08-29 · 30b83f018334
- site_page: https://xerj.org/docs/recipes/zero-config-autoindex · fetched 2026-08-29 · 8beaee837c61
- site_page: https://xerj.org/docs/recipes/agentic-memory · fetched 2026-08-29 · e4286298b8ca
- site_page: https://xerj.org/docs/quickstart · fetched 2026-08-29 · b799e5a56677
- site_page: https://xerj.org/docs/install · fetched 2026-08-29 · 682910f8f385
- site_page: https://xerj.org/docs/migration-from-es · fetched 2026-08-29 · e60fc8ab829e
- site_page: https://xerj.org/pricing · fetched 2026-08-29 · 1fbc3ba5018d
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| xerj-org/xerj | main | 77 |
For agents
Data as of 2026-10-19T08:05:51.212177+00:00 · Report a problem