open-timeline-engine vs deja-vu
Local memory tools for coding agents compared
open-timeline-engine and deja-vu help coding agents reuse context from earlier work through local memory and MCP. open-timeline-engine captures activity and decisions over time, while deja-vu indexes existing agent session histories for search and recall.
open-timeline-engine: Shared Memory for AI Agents via MCP
Open Timeline Engine is a local-first Python context engine for MCP-compatible agents such as Claude, Codex, and Cursor. It captures work over time, finds recurring patterns, and returns past decisions with citations and safety gates.
deja-vu: Search Coding-Agent History for Reusable Memory
deja-vu is a Go-based local memory tool that indexes existing session history from Claude Code, Codex, Cursor and 39 more coding agents. It provides search, MCP recall and hooks without requiring an LLM, so agents can reuse earlier decisions and fixes.
| open-timeline-engine | deja-vu | |
|---|---|---|
| Language | Python | Go |
| License | Apache-2.0 | MIT |
| Stars | 9 | 1.2k |
| Forks | 0 | 128 |
| Last analyzed | Oct 15, 2026 | Oct 19, 2026 |
Key differences
- open-timeline-engine captures timeline events from agent sessions and sources such as Git, editors, and browsers; deja-vu indexes session history already stored on disk.
- open-timeline-engine combines timeline, pattern, and graph data with citations and policy controls; deja-vu focuses on searchable history, MCP recall, hooks, and commands for inspecting prior work.
- open-timeline-engine is written in Python and licensed under Apache-2.0; deja-vu is written in Go and licensed under MIT.
- open-timeline-engine describes itself as experimental and not production-ready; deja-vu describes macOS and Linux as its more established platforms and asks for Windows field reports.
- open-timeline-engine supports shared or per-executor workspaces and includes a dashboard; deja-vu supports session handoff, curated notes, and moving an index between machines.
Choose open-timeline-engine if you…
- need to capture activity from agent sessions plus Git, editor, or browser sources.
- want cited context, advisor mode, and layered controls for agent actions.
- need shared or separate memory across executors and are prepared to evaluate an experimental project.
Choose deja-vu if you…
- want to search existing histories from Claude Code, Codex, Cursor, and other coding agents.
- need CLI or terminal search, MCP recall, or hooks without requiring an LLM.
- want session handoff, history-inspection commands, or optional semantic search through an embeddings service.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.