Back to browse
GitHub Repository

Local-first context engine that gives AI agents persistent, shared memory. Captures your real workflow, mines repeatable patterns, and serves past decisions back to agents like Claude, Codex, and Cursor — with citations, safety gates, and your learned working style enforced.

7 starsPython

OpenTimelineEngine – Shared local memory for Claude Code and codex

by joeljoseph_·Mar 2, 2026·3 points·1 comment

AI Analysis

MidBig BrainShip It

AI memory layer with audit trails, but pre-1.0 and lacks live integrations.

Strengths
  • Genuine problem framing — AI agents forgetting context across sessions is real friction.
  • Safety-first design: permit-claim-execute-report lifecycle prevents blind agent actions.
  • Shared memory across executors means agents learn from each other's corrections.
Weaknesses
  • Pre-beta experimental status (v0.3.0) signals APIs and storage formats will break.
  • No live integrations shown for Claude, Cursor, Codex — concept isn't wired to real agents yet.
  • Crowded space: long-context windows (Claude 200k), agentic frameworks (Langgraph), and persistent memory tools (Mem0, AnythingLLM) overlap heavily.
Target Audience

Solo and repeat users of AI coding agents (Claude, Cursor, Codex) who want persistent context and auditability across sessions.

Similar To

Mem0 · AnythingLLM · LlamaIndex

Similar Projects

Developer Tools●●Solid

OpenTimelineEngine – Shared local memory for Claude Code and codex

Local memory for AI agents that actually learns from your last 50 sessions, not just context window tricks.

Big BrainShip ItSolve My Problem
joeljoseph_
103mo ago

OpenTimelineEngine – Shared local memory for Claude Code and codex

Persistent memory for Claude across sessions, but API stability unclear at v0.3.0 beta.

Big BrainShip It
joeljoseph_
523mo ago