Item detail

aayoawoyemi/Ori-Mnemos

aayoawoyemi/Ori-Mnemos is an Apache-2.0-licensed, local-first persistent memory infrastructure for AI agents that implements human cognition as mathematical models on a markdown-backed knowledge graph with activation decay, spreading activation, Hebbian co-occurrence, and recursive graph traversal so agents can remember across sessions without a vendor cloud or database lock-in.

Score7.8
Popularity7.6
Risklow
TierSilver
Score breakdown
Usefulness8.0
Novelty9.0
Momentum7.0
Maturity5.8
Open-source/build8.4
Evidence7.2
Workflow potential9.3
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for AI agent builders, researchers, and self-hosters who want a local-first, markdown-on-disk persistent memory layer for AI agents with no database lock-in, so agents can remember, forget, and re-learn across sessions without uploading the knowledge graph to a vendor cloud or depending on a hosted memory SaaS.

Who should use it

AI agent builders who want a local-first persistent memory layer for agents with no vendor lock-inresearchers who need an inspectable markdown-on-disk knowledge graph for agent memory experimentsself-hosters who want Git-versioned agent memory instead of a hosted vector databaseopen-source contributors who want an Apache-2.0 alternative to closed-source agent memory SaaS

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

It is a local-first persistent memory layer where every byte of the user's knowledge graph lives on disk in markdown, so audit what is committed to the knowledge graph, review the activation decay + spreading activation parameters for any sensitive content before enabling auto-retrieval, and back up the markdown knowledge graph before any schema-level migration.

Evidence links

Closest alternatives / related signals

agent-memoryknowledge-graphlocal-firstmarkdownopen-sourceapache-2.0