Fraise’s creator claims its single-binary temporal-graph database provides ranked, capped agent-memory retrieval through remember and recall commands, potentially bounding retrieved context while simplifying persistent-memory infrastructure.
state: expiredheat: lowuncertainty: highknownscott: lowagent-memory temporal-knowledge retrievalFraiseairejie
What is this?
The supplied case describes Fraise as an AI-agent memory database presented in a Show HN post, associating its creator with the handle airejie. Its creator reportedly claims a single-binary temporal-graph design with remember and recall commands that return ranked, capped memories. None of the supplied search results directly documents Fraise or verifies those features; they discuss graph-based agent memory and temporal retrieval generally, so simpler infrastructure and bounded retrieved context remain proposed benefits rather than demonstrated outcomes.
Why it matters to Scott
Fraise’s claimed external memory and capped retrieval repeat positions already held in Scott’s long-running-agents and context-engineering pages; no supplied radar hit tracks Fraise itself. The unverified creator claims establish neither retrieval quality nor operational savings, so this is another example of his existing architecture rather than evidence to change his builds or arguments.
ip:framework.long-running-agentsip:framework.context-engineeringradar:concept.agent-memoryradar:concept.retrievalradar:engrava-local-graph-agent-memoryradar:polign-stateless-agent-memory
queries asked of Scott's wikis
- persistent agent memory architecture and deployment complexity
- ranked retrieval context budgets and token limits
- temporal knowledge graphs changing facts and memory provenance
- agent-maintained wikis versus structured graph memory
- memory retrieval quality evaluation and recall tradeoffs
Measured heat
no measured readings yet — the hourly heat pass fills this in
How the heat travelled
no chain yet — the hourly chain pass fills this in
Evidence (1) — ⭐ canonical anchor
Interpretation history
2026-09-10T13:31:15Z
The stale recheck supplies no new evidence beyond the creator’s announcement, leaving Fraise an implementation lead rather than demonstrated progress in memory retrieval or deployment simplicity. With no pending validation or expected milestone identified, retire active monitoring without treating the claims as disproved.
2026-09-08T12:32:59Z
No substantive new evidence changes Fraise’s status as a creator-announced memory tool rather than a demonstrated improvement in retrieval quality or deployment simplicity. Retain it as an implementation lead, without importing momentum from the broader agent-memory topic.
2026-09-08T12:29:06Z
grounded: known/low — Fraise’s claimed external memory and capped retrieval repeat positions already held in Scott’s long-running-agents and context-engineering pages; no supplied ra
2026-09-08T12:25:39Z
case created — The first-party launch describes a concrete storage and retrieval design with linked documentation, distinct from the existing memory-tool episodes.
Decision trace
- 09-10 23:31expireThe stale recheck supplies no new evidence beyond the creator’s announcement, leaving Fraise an implementation lead rather than demonstrated progress in memory retrieval or deployment simplicity. With
- 09-10 23:31alert_silentThere is no new release, implementation evidence, or comparative result to surface. The existing announcement can wait for a tooling briefing; independent usage or retrieval evaluations would justify
- 09-10 23:31alert_routeThere is no new release, implementation evidence, or comparative result to surface. The existing announcement can wait for a tooling briefing; independent usage or retrieval evaluations would justify
- 09-08 22:32repriceNo substantive new evidence changes Fraise’s status as a creator-announced memory tool rather than a demonstrated improvement in retrieval quality or deployment simplicity. Retain it as an implementat
- 09-08 22:32alert_silentThis recheck adds no consequential delta to the already assessed announcement. The offering can wait for a tooling briefing; independent usage or comparative results would provide a new reason to revi
- 09-08 22:32alert_routeThis recheck adds no consequential delta to the already assessed announcement. The offering can wait for a tooling briefing; independent usage or comparative results would provide a new reason to revi
- 09-08 22:31alert_silentThe creator’s announcement establishes a new tool offering: temporal-graph memory, ranked and capped retrieval, and integration with Claude Code and Codex. Retrieval quality and operational savings re
- 09-08 22:31alert_routeThe creator’s announcement establishes a new tool offering: temporal-graph memory, ranked and capped retrieval, and integration with Claude Code and Codex. Retrieval quality and operational savings re
- 09-08 22:29groundFraise’s claimed external memory and capped retrieval repeat positions already held in Scott’s long-running-agents and context-engineering pages; no supplied radar hit tracks Fraise itself. The unveri
- 09-08 22:25createThe first-party launch describes a concrete storage and retrieval design with linked documentation, distinct from the existing memory-tool episodes.