memory-bank

Manage shared git-based memory banks with sync, collision checks, and pull-request contributions.

Updated May 3, 2026
One-click install
npx skills add https://github.com/spikelab/multiplai-cc-mktplace --skill memory-bank-spikelab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-bank
Source: https://github.com/spikelab/multiplai-cc-mktplace/tree/main/plugins/multiplai-context/skills/memory-bank
Command: npx skills add https://github.com/spikelab/multiplai-cc-mktplace --skill memory-bank-spikelab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Teams and households need to share curated memory across multiple Claude Code setups without duplicating facts or letting shared content silently overwrite personal memory. This Skill manages git-repository memory banks so a fact lives in exactly one authoritative place. ## Core Features & Use Cases - Bank subscription and sync: List configured banks, fast-forward sync them via git, and rebuild routing catalogs so shared memory is indistinguishable from personal memory. - Collision detection and contribution: Report cross-bank duplication as defects, and turn dream-proposal items into a branch, commit, and pull request on a bank with no-go domain and credential scanning. - Adopt migration: Move personal memory into a bank by deleting local copies only after the content is verified line-for-line in the bank, with a revert receipt for every deletion. - Use Case: A team maintains a shared memory bank of deployment procedures. A member subscribes, runs a collision check, contributes a new fact via pull request, and after merge adopts the bank version, deleting their local duplicate safely. ## Quick Start Ask the assistant to list my configured memory banks and sync them, then run a collision check across all banks.

Frequently Asked Questions about memory-bank

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I share memory across a team using git repositories?

Create a memory bank as a git repository of memory files, then have each member declare it in memory-banks.yaml with a remote, mode, and sync policy. Rebuild the catalogs and the bank's files route exactly like personal memory.

How do I contribute memory to a shared bank?

Run the contribute verb with a dream proposal path to create a branch, commit, and pull request on the bank. Items are checked against the bank's no-go domains and scanned for credential shapes before the PR opens.

Can a shared memory bank be written to directly?

No, shared banks have no read-write mode; configurations requesting one are coerced to propose mode. All changes leave as pull requests, and the write floor refuses any local write targeting a non-personal bank.

What happens if my personal memory duplicates bank content?

The check command reports cross-bank collisions as defects since a fact should live in exactly one bank. The adopt migration resolves this by deleting local files only after verifying the content exists line-for-line in the bank.

Does the memory bank skill require the GitHub CLI?

The gh CLI is only needed for the final pull request creation step of contribute. Without it, the branch is still pushed and you are told to open the pull request manually; git alone covers everything else.