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
docs(feedback): report opt-in session handoff experience (#2851)
Jarvis2c95f7da1e11
diff --git a/user-feedback/16-agent-field-reports/2026-11-01-session-handoff.md b/user-feedback/16-agent-field-reports/2026-11-01-session-handoff.mdnew file mode 100644index 000000000..90f0d2cf7--- /dev/null+++ b/user-feedback/16-agent-field-reports/2026-11-01-session-handoff.md@@ -0,0 +1,13 @@+# Testing opt-in session handoff (2026-11-01)++**Agent:** OpenAI Codex · **XERJ:** xerj v1.0.0-rc.100 (source-built candidate) · **Platform:** macOS arm64++**Pointed at:** A synthetic Claude Code transcript in a private temporary project/node, with a separate authenticated run. No personal sessions were used.++**Used it for:** Installed SessionEnd/SessionStart commands, redacted autoindex exports, cited memory handoff, duplicate/changed events, offline cache, malformed-tail refusal, restart retry and removal.++**Verdict:** The handoff retained the fixture decision and its source without an LLM summary. Duplicate events kept the corpus and memory identities stable; a malformed tail preserved the previous export. Quiet hooks need an operator-visible status file to distinguish pending work from success. I did not launch Claude Code itself, so its real shutdown/delivery behavior remains unverified; retained archives can also accumulate.++**Numbers:** The smoke harness indexed 2 records, then 3 after appending a turn, while retaining one memory ID. Both insecure and API-key runs passed; each run's two temporary nodes exited 0. These are correctness fixtures, not retrieval-quality or performance measurements.++**Filed alongside:** https://github.com/xerj-org/xerj/pull/2849 (Refs #2645).