Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors
Rust
test(mcp): wait for the async burst fallback
AkitaOnRails1ab58c0dc6ac
diff --git a/crates/ai-memory-mcp/tests/suite/stress_autoscope.rs b/crates/ai-memory-mcp/tests/suite/stress_autoscope.rsindex 4f0ac47b..c3e5074b 100644--- a/crates/ai-memory-mcp/tests/suite/stress_autoscope.rs+++ b/crates/ai-memory-mcp/tests/suite/stress_autoscope.rs@@ -683,11 +683,17 @@ async fn burst_above_max_entries_does_not_corrupt() {h.await.expect("burst hook join");}- // Settle: every spawned process_envelope must drain.- for _ in 0..50 {+ // The hook response is deliberately fire-and-forget: 202 can arrive before+ // any spawned `process_envelope` publishes the shared fallback. Poll the+ // invariant this test later relies on instead of racing a fixed sleep on+ // slower Windows runners.+ for _ in 0..200 {+ if h.active_project.get().is_some() {+ break;+ }tokio::task::yield_now().await;+ tokio::time::sleep(Duration::from_millis(5)).await;}- tokio::time::sleep(Duration::from_millis(50)).await;// Now BURST_SIZE parallel reads. Every read must return a JSON-RPC// 200 — no panic, no 500. The CONTENT may be any of the projects