Skip to content

Concepts

agent-memory pairs a deterministic memory substrate with a lightweight cognitive loop. Together they turn memory-aware agent work into predictable innovation with human partnership — bold ideas, faithful delivery, a human in the loop at every altitude.

  • Evolving Memory


    The backward layer: an event-sourced ledger projected into live state, with deterministic decay, supersession, and provenance. Faithful to what happened.

  • The VBDI Loop


    The forward layer: Current State → Vision → Blueprint → Design → Implementation → Feedback, with an enforceable intent trace. Faithful to what was intended.

  • Skills & Built-ins


    Reusable capabilities authored once in a neutral agent-skills/ layer and run by any agent — plus seven tool-managed built-ins.

  • Decay & Tiers


    How facts strengthen, fade, and archive — by counting session files, never by a floating-point score.

  • Reliability


    How the ritual actually happens without depending on an agent self-triggering: triggers, CI, init, and the judgment-vs-arithmetic boundary.

The mental model

There are two distinct memory layers in a developer's life, and agent-memory keeps them separate by design:

Layer Where it lives Who owns it What it holds
Personal ~/ (home) You Your vendor's chat history, model prefs, global settings
Shared (team) the repo's memory/ The team, via git Project rules, decisions, session logs

The tool only ever touches the team layer. Your personal ~/.claude/, ~/.cursor/, Application Support, AppData — none of it is read, modified, or moved. Whatever vendor you prefer keeps working exactly as before.