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
fix(ci): the toolkit test step never ran — one filter per invocation, and xc_ matched nothing
xerj-org3aba5786497b
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.ymlindex fdbd4fa59..3b3bf69da 100644--- a/.github/workflows/ci.yml+++ b/.github/workflows/ci.yml@@ -104,8 +104,16 @@ jobs:- name: Toolkit testsworking-directory: enginerun: |- cargo test --profile ci-test -p xerj-common -p xerj-autoindex -p xerj-mcp -p xerj-server \- xccode xc_ published_schema help_text+ # cargo test takes ONE name filter per invocation — the four+ # filters are separate runs (the single-command form never parsed:+ # "unexpected argument 'xc_' found", and the branch run that would+ # have caught it was cancelled before this step). xc::, not xc_:+ # the autoindex tests live at xc::tests::… and no test name ever+ # contained the substring "xc_" — that filter matched zero tests+ # from the day it was written.+ for filter in xccode xc:: published_schema help_text; do+ cargo test --profile ci-test -p xerj-common -p xerj-autoindex -p xerj-mcp -p xerj-server "$filter"+ done# The corpus manifests are the shareable artifact (issue #319): if they# cannot rebuild a corpus, publishing them is worse than not shipping