Memory Bank

A durable, version-controlled context and governance layer for software development with coding agents.

Русская версия · Adoption guide · Daily usage · CLI

Coding agents are most useful when they share the same understanding of the product, domain, architecture, constraints, and definition of done. Memory Bank keeps that knowledge in Git, next to the code, instead of leaving it in one person's head or a disposable chat session.

It gives humans and agents an authoritative starting point, routes work through explicit delivery flows, and preserves the decisions needed to resume a task in a fresh session.

Memory Bank is not a wiki, task tracker, or agent runner. It is the control plane around those tools: durable context, ownership rules, lifecycle gates, and verification contracts.

• Durable project context — product intent, domain language, engineering rules, and operational constraints survive across sessions.
• Clear ownership — Single Source of Truth rules prevent the same fact from drifting across documents.
• Governed delivery — task routing selects the smallest suitable flow for incidents, bugs, research, small changes, epics, refactoring, or features.
• Safe reuse and updates — an ownership-aware CLI installs the template, preserves local customizations, and reports conflicts.
• Automated checks — lint audits links and navigation; doctor checks adoption, governance, managed drift, and CI integration.

Memory Bank context and rules
             ↓
         Issue / task
             ↓
   Agent session and delivery flow
             ↓
 Implementation → verification → PR
             ↓
 New durable knowledge returns to Memory Bank

Install the companion CLI with Go:

go install github.com/dapi/memory-bank-cli/cmd/memory-bank-cli@latest

Then, from the root of your Git repository:

memory-bank-cli init --dry-run
memory-bank-cli init
memory-bank-cli doctor

--dry-run previews every file before installation. init installs the template, creates memory-bank/.lock, and adds a managed Memory Bank block to the repository's agent instructions. For reproducible automation, pin a released CLI version instead of latest.

Next, open memory-bank/README.md and adapt product/, domain/, engineering/, and ops/ to the actual project. Generic template text is not evidence about an existing codebase: brownfield projects should follow the brownfield adaptation protocol, while new projects can use the greenfield protocol.

See the complete adoption guide for agent setup, local validation, and downstream CI.

The dna/ layer defines document governance: source ownership, dependency direction, lifecycle, frontmatter, and navigation. Stable project context lives in product/, domain/, engineering/, and ops/. Requirements and decisions mature through research, PRDs, epics, use cases, feature packages, and ADRs.

For a substantial delivery feature, the context typically develops in three stages:

brief.md                 design.md                  implementation-plan.md
what and why      →      chosen solution     →     implementation and checks
problem space            solution space             execution space
                         (when required)

Documents own intent, requirements, rationale, and contracts. Code owns implementation. A new agent session can therefore restart from the same task and canonical documents without reconstructing the project from chat history.

Every task begins with Task Routing, which selects the applicable lifecycle and its evidence requirements.

This repository is the upstream source. memory-bank-cli init installs tracked regular files from template/ into a downstream repository: template/memory-bank/ becomes memory-bank/, while template/init.sh becomes ./init.sh.

After installation, memory-bank/README.md is the primary index inside the downstream project.

• Adopting Memory Bank
• Using Memory Bank day to day
• Context priming for an agent task
• CLI integration
• Ownership and safe updates
• Repository development
• Detailed overview in Russian

The governance model applies the MECE principle: categories should be mutually exclusive and collectively exhaustive within their declared scope.

The CLI is developed and released separately in dapi/memory-bank-cli. This template is available under the Apache License 2.0.