fluctlightdb 0.5.1

Embedded memory engine for AI agents — third data model (experience/activate), provenance, benchmarks. Goal: SQLite for agent memory.

FluctlightDB

The memory engine for AI agents — not a vector database with an agent SDK bolted on.

· GitHub

Mission

Goal: become the default database for agent memory — the way SQLite became the default embedded DB for apps.

Long-term agent memory is a third data model (alongside relational facts and vector similarity). FluctlightDB defines engine-level experience() / activate() semantics — episodes, cue-driven recall, provenance, consolidation — not app glue on top of Chroma or Mem0.

Who it's for — agents that persist across sessions, learn from tools/files/APIs as well as chat, recall under paraphrase, and should prefer verified evidence over casual conversation at recall time.

Typical fits: coding agents, ops bots, research assistants, NPCs.

Install

pip install "fluctlightdb[native]"

from fluctlightdb import connect

brain = connect("/tmp/my-agent-brain")
brain.experience("User prefers dark mode", context="settings", salience=0.8)
print(brain.activate("theme preference"))
brain.checkpoint()

HTTP-only (no Rust extension): pip install fluctlightdb

Benchmarks (June 2025)

Benchmark	Metric	Result
LoCoMo (10 conv)	Mean evidence recall @ k=150	98.1%
BEIR SciFact	nDCG@10 (index mode)	0.645 (ties Chroma + MiniLM)
FAMB	Macro (index / agent)	98% / 97%

Frozen JSON: benchmarks/results/2025-06-22.json

LoCoMo evidence recall ≠ Mem0 LLM-as-judge QA — different metrics; compare only when labeled.

Docs

• Getting started
• Multi-agent monorepos — fluctlight-project init, MCP, hooks, handoffs
• Platform compatibility — Windows, macOS, Linux
• Full README & reproduction
• Platform checklist