🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
TypeScript agentic-aiagentic-frameworkagentic-workflowagentsai-agentsai-assistantai-skillsautonomous-agentsclaudecodexdsh-pluginharnessmcp-servermulti-agentmulti-agent-systemsnpmskillsswarmswarm-intelligencetypescript
test(memory): the #3374 'well-formed query reaches the search layer' guard allows the one near-miss probe #4129 adds
r9d9f490b8f52
diff --git a/v3/@claude-flow/cli/__tests__/issue-3374-memory-missing-required-param.test.ts b/v3/@claude-flow/cli/__tests__/issue-3374-memory-missing-required-param.test.tsindex 3bbac34db5..2642083ce8 100644--- a/v3/@claude-flow/cli/__tests__/issue-3374-memory-missing-required-param.test.ts+++ b/v3/@claude-flow/cli/__tests__/issue-3374-memory-missing-required-param.test.ts@@ -89,7 +89,14 @@ describe('#3374 memory tools reject a missing required parameter at the boundaryit('a well-formed query still reaches the search layer', async () => {const r = (await tool('memory_search').handler({ query: 'cross-agent proof bridge', limit: 5 })) as Res;- expect(layer.searchEntries).toHaveBeenCalledTimes(1);+ // The real search runs first with the caller's query. Since #4129 an EMPTY default-threshold search+ // may be followed by exactly one near-miss probe (threshold 0) for the hint; nothing else.+ expect(layer.searchEntries.mock.calls[0]?.[0]).toMatchObject({ query: 'cross-agent proof bridge' });+ expect(layer.searchEntries.mock.calls[0]?.[0]).not.toHaveProperty('threshold', 0);+ expect(layer.searchEntries.mock.calls.length).toBeLessThanOrEqual(2);+ if (layer.searchEntries.mock.calls.length === 2) {+ expect(layer.searchEntries.mock.calls[1]?.[0]).toMatchObject({ query: 'cross-agent proof bridge', threshold: 0 });+ }expect(r.code).toBeUndefined();expect(r.error).toBeUndefined();});