Ultimate Memory

Project Status

Ultimate Memory is being built design-first: the complete system — requirements, architecture, data model, and every load-bearing decision with its rationale — was designed and reviewed before implementation began. That design is public in the repository and is the authority these docs summarize.

There is no installable release yet. If you need a working memory library today, this isn't one — yet. If you want to follow (or join) a system being built in the open against a fully argued design, read on.

What exists today

  • The complete design — in plan/: requirements, binding per-subsystem designs, the research and analyses behind them, and a build-order roadmap.
  • The decision logdecisions.md: every architecture decision (D1, D2, …) with context, rationale, and its later refinements. If you want to know why something is the way it is, start there.
  • Design-conformance evals — a suite of acceptance checks (in plan/implementation_evals/) that judge the implementation against the binding design as it lands.
  • Early scaffolding — the Python project skeleton and CI are in place; pipeline functionality is landing phase by phase.

The build order

Implementation follows plan/plans/roadmap.md — nine phases, each keyed to the designs it realizes:

PhaseWhat it delivers
0Foundations: scaffolding, migrations, task queues, the eval harness
1Walking skeleton — one document end-to-end, minimal retrieval
2Truth machinery — entity resolution, adjudication, supersession
3Evidence lifecycle — document versions, watched sources, efficient re-ingestion
4Projections — the knowledge graph and the browsable corpus filesystem
5Retrieval complete — full query surface, MCP/CLI, the consumption skill
6Plane K — compiled and authored knowledge pages
7Scale and operations — backfill, load tests, budgets, hard-delete end-to-end
8Competitive benchmarks — published methodology and results

How these docs grow

These pages are part of the repository and ship with the code: every change that alters user-facing behavior updates its documentation in the same pull request. As each phase lands, its pages appear here — getting started, ingestion and connectors, retrieval, configuration, deployment, and the API/CLI/MCP references. Until a page exists, the design documents linked above are the truth.

Following along

  • Watch the GitHub repository — work lands as reviewed pull requests referencing roadmap work packages.
  • Read the decision log for the ideas; skim merged PRs for the build.