Memory Bank
Persistent .memory-bank/ project memory, TDD/Clean-Architecture rules, and a 25+ command dev toolkit.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 15, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 53a0ca7
Copied the raw repo straight from GitHub with no installer, and `mb-init-bank.sh` / `mb-note.sh` / `mb-idea.sh` / `mb-search.sh` all worked immediately on a throwaway project, scaffolding a real structured `.memory-bank/` (status, roadmap, backlog with monotonic I-NNN ids, timestamped notes) that a plain 'write some notes' baseline can't match. README's '1,900+ automated tests' claim is actually conservative — repo has 2,275 bats + 1,542 pytest tests, and the 29-subagent count matches exactly.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 9/10
- Docs & honesty 5/5
What Memory Bank does
Gives agents durable, git-tracked project memory in `.memory-bank/` (status/plan/checklist/backlog/lessons) plus an enforced engineering-rules baseline (TDD, SOLID, Clean Architecture, FSD) and 25+ `/mb` workflow commands. Triggers automatically in any project with a `.memory-bank/` directory, or when the user explicitly asks for a memory-bank workflow, engineering rules, or dev-toolkit commands.
How to install Memory Bank
git clone https://github.com/fockus/skill-memory-bank
mkdir -p ~/.claude/skills
cp -r skill-memory-bank ~/.claude/skills/memory-bank
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Memory Bank
-
/memory-bankPersistent .memory-bank/ project memory, TDD/Clean-Architecture rules, and a 25+ command dev toolkit.
It also activates on plain-language prompts like these:
-
Set up persistent project notes so agents remember context each time. -
This project already has saved status notes, let's resume from there. -
Log this feature idea into our project's backlog for later.
Frequently asked questions
- Is the Memory Bank skill free?
- Yes. The skill itself is free from fockus/skill-memory-bank. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Memory Bank work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 15, 2026. Verdict: Tested · Works. Copied the raw repo straight from GitHub with no installer, and `mb-init-bank.sh` / `mb-note.sh` / `mb-idea.sh` / `mb-search.sh` all worked immediately on a throwaway project, scaffolding a real structured `.memory-bank/` (status, roadmap, backlog with monotonic I-NNN ids, timestamped notes) that a plain 'write some notes' baseline can't match. README's '1,900+ automated tests' claim is actually conservative — repo has 2,275 bats + 1,542 pytest tests, and the 29-subagent count matches exactly.
- What is the Memory Bank SkillProof Score?
- 9.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 9/10, docs & honesty 5/5.
- How do I install Memory Bank?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Memory Bank with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.