Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/cwakim/claude-pluginsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/plugins/cwakim/claude-plugins/memory-backup)<a href="https://agentmods.dev/plugins/cwakim/claude-plugins/memory-backup"><img src="https://agentmods.dev/badge/plugins/cwakim/claude-plugins/memory-backup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/plugins/cwakim/claude-plugins/memory-backup"><img src="https://agentmods.dev/badge/plugins/cwakim/claude-plugins/memory-backup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
memory-backup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 23d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
92% identical to cwakim-plugins marketplace — 37 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 10 lines — stays where its author put it; the contents beside it link to each section on GitHub.
{
"name": "memory-backup",
"description": "Back up Claude Code's machine-local state (/backup): mirror every per-project memory store, handoff note (live and archived), plan document, and the hand-written global config to a private GitHub repo, one squash-merged PR per run. Repo visibility verified before every push, refused if public; every mirrored file secret-scanned before commit (interactive runs ask, cron redacts and flags loudly), credentials never leave the machine. Diff-aware restore fills empty targets wholesale and asks per conflicting file, from GitHub or from a zip, confirming the plan before anything is applied (--dry-run previews the plan and touches nothing). merge converges any number of machines' memories, handoffs, and plans (one, several, or all other mirrors per run) with the same temperament plus keep-both conflict resolution, cross-machine path reconciliation, and union-rebuilt indexes; interactive-only. zip <path> exports the same mirror as a portable dated archive with no GitHub required, asking each run whether to secret-scan first. share <thread> builds a curated, sanitized handoff bundle for one thread (its anchor handoff note, the memory stores it touches with project and reference memories only, the plans you pick, a generated ONBOARDING.md), secret-scanned every run and delivered as a secret gist or a zip, so a colleague with no repo access can pick the work up with their own restore or merge; interactive-only. Optional schedule for the backup (cron, or launchd on macOS so runs missed while asleep fire on wake), cadence is your call (weekly default). An optional S3-compatible object-storage target (AWS S3, GCS, Alibaba OSS, or a self-hosted MinIO) can back up the same tree to a bucket alongside or instead of GitHub: setup creates and hardens the bucket (versioning on, refuses a public bucket), backup mirrors with delete-propagation and the same secret scan, and restore and merge read a bucket the same way they read the repo; boWhat it installs
The manifest is a name and a version. 5 commands travel with it, and installing the plugin installs all of them — 332 tokens a session between them. Each is measured on its own page, and each can be installed alone.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 23d ago Changed 23530bb39629
- 1mo ago Changed c5a27ff26702
- 1mo ago First seen · 10 lines scan A 60ce06fad67c
memory-backup is a plugin published in the GitHub repository cwakim/claude-plugins (2 stars, last pushed 23d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. It is 92% identical to cwakim-plugins marketplace, differing in 37 lines, and is treated as a copy.
Other plugins, from other repositories
yantrikdb
Cognitive memory for AI agents — semantic recall, a knowledge graph, contradiction detection and procedural learning. Local-first: one SQLite file, no account, no telemetry.
cc-memory
Persistent memory for Claude Code — automatic save/restore across compactions and sessions. Anti-patch reconcile-on-write + LLM-judged semantic de-duplication, forced PROGRESS.md handoff with per-session annotation, live PLAN.md anchor, injection observability, FTS5 search, AI-judged extraction with Haiku + local…
presence
Continuous-collaboration layer for Claude Code: living project model, outcome telemetry, event digest, and calibrated confidence. Fully local, install-once, applies to every project.
core-memory-kit
Per-project in-repo memory system for Claude Code. Frozen-snapshot session loads, auto-extract Stop hook, memory-write + memory-search + troubleshooting skills, keyword search, optional auto-curation crons.
jarvis-os
A persistent memory hub for Claude Code that loads into every session from any directory, plus a curated set of research and workflow skills.
claimkeep
Continuous memory that survives context compaction — harvests your agent's own confidence-marked claims into a verbatim brief and re-injects it after compaction.