Ruflo v3.42.4 — smart search score semantics fix: What changed Fixes Finding B of #3327: memory_search({ smart: true }) was reporting SmartRet
The official v3.42.4 — smart search score semantics fix release documents a concrete integrations change. This update explains what changed from v3.42.3 — Windows publish + hook-override quoting fixes, who should reassess the product, and what to verify next.
- What changed
- The official v3.42.4 — smart search score semantics fix release documents a concrete integrations change. This update explains what changed from v3.42.3 — Windows publish + hook-override quoting fixes, who should reassess the product, and what to verify next.
- Who is affected
- Evaluators, administrators, and workflow owners using Ruflo where integrations requirements affect adoption or deployment.
- What to do now
- Review the official v3.42.4 — smart search score semantics fix notes, test the affected integrations workflow, and update the adoption decision if the result changes product fit.
The material change
Previously: {{agentName}} v3.38.21 — MCP HTTP bridge memory-persistence fix documents this product change: Fixed #3155 — ruflo mcp start -t http: entries stored via the bridge were unreadable after a bridge restart ( found:false, memory_list empty), even though.swarm/memory.db held them and the CLI read them fine.
Now: Ruflo v3.42.4 — smart search score semantics fix documents this product change: What changed Fixes Finding B of #3327: memory_search({ smart: true }) was reporting SmartRetrieval's internal RRF fusion score in the similarity field instead of the underlying retrieval relevance score — e.g.
This update covers the v3.42.4 — smart search score semantics fix release notes and does not infer behavior beyond the official release description.
What users should reconsider
The move from v3.42.3 — Windows publish + hook-override quoting fixes to v3.42.4 — smart search score semantics fix introduces a documented change that may alter operational compatibility.
The release has a versioned first-party source, so teams can verify the exact scope instead of relying on a generic monitoring excerpt.
Verify the affected workflow
- Review the official v3.42.4 — smart search score semantics fix notes, test the affected integrations workflow, and update the adoption decision if the result changes product fit.
- Open the official source and verify the current product state.
- Review the current Agent Profile and affected Decision before acting.
Source-Backed does not mean execution-verified and does not establish comparative rank.
What was recorded before
{{agentName}} v3.38.21 — MCP HTTP bridge memory-persistence fix documents this product change: Fixed #3155 — ruflo mcp start -t http: entries stored via the bridge were unreadable after a bridge restart ( found:false, memory_list empty), even though.swarm/memory.db held them and the CLI read them fine.
What is known now
Ruflo v3.42.4 — smart search score semantics fix documents this product change: What changed Fixes Finding B of #3327: memory_search({ smart: true }) was reporting SmartRetrieval's internal RRF fusion score in the similarity field instead of the underlying retrieval relevance score — e.g. This update covers the v3.42.4 — smart search score semantics fix release notes and does not infer behavior beyond the official release description.
Accepted Change accepted:499f96cfe7c6ce55bf31ab37d3521a41 · Agent Fact fact:499f96cfe7c6ce55bf31ab37d3521a41 · Last verified 2026-09-17Ruflo official releases
Published by Ruvnet. Evidence state remains Source-Backed; this is not execution verification.
Open official source →