agentmods.dev

memory-save

memory-save is a command for Claude Code from gaebalai/claude-memory-plugin. It costs 15 tokens per session (837 once invoked), scanned A, original, MIT.

A session-memory command that classifies information as a user preference, feedback, project update, or external reference, then prepares a memory file for it.

In plain words
What is it for?
Use it to sort notes into memory files, suggest filenames and required fields, and update the memory index.
Why use it?
It helps keep useful session knowledge organized instead of leaving it scattered or forgetting where it belongs. It also requires approval before saving.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

Part of the memory-system plugin — 4 commands shipped together

not rated 2repo 6mo ago A scan 15 tokens original MIT

Good fit Use it to sort notes into memory files, suggest filenames and required fields, and update the memory index.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/gaebalai/claude-memory-plugin/memory-save
View source ↗ gaebalai/claude-memory-plugin
Install

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.

Clone the repo
git clone --depth 1 https://github.com/gaebalai/claude-memory-plugin

Made for: Claude Code.

Or install memory-system, the plugin that ships this one along with the rest of its 4 commands.

Wrote 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.

agentmods badge for memory-save

README.md
[![agentmods](https://agentmods.dev/badge/commands/gaebalai/claude-memory-plugin/memory-save/github.svg)](https://agentmods.dev/commands/gaebalai/claude-memory-plugin/memory-save)
Your own site
<a href="https://agentmods.dev/commands/gaebalai/claude-memory-plugin/memory-save"><img src="https://agentmods.dev/badge/commands/gaebalai/claude-memory-plugin/memory-save/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.

agentmods 80×15 button for memory-save

Your own site · 80×15
<a href="https://agentmods.dev/commands/gaebalai/claude-memory-plugin/memory-save"><img src="https://agentmods.dev/badge/commands/gaebalai/claude-memory-plugin/memory-save.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 837 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00015 $0.00837
Opus 5.5 $0.00006 $0.00335
Sonnet 5.5 $0.00003 $0.00167
Haiku 4.5 $0.00002 $0.00084

Measured 2mo ago against content hash 463aa8ca7960, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

memory-save 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 2mo 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.

plugins/memory-system/commands/memory-save.md · 85 lines

What it actually says

Memory 저장

다음 내용을 읽고 가장 적합한 memory 타입으로 분류하여 저장합니다:

내용: $ARGUMENTS

분류 기준

타입 언제 사용 파일명 규칙 필수 필드
user 사용자 선호·스킬·직책 user_*.md name, type, description
feedback 과거 지적받거나 배운 행동 규칙 feedback_*.md Why + How to apply 필수
project 프로젝트의 동적 상태 project_*.md 날짜 (YYYY-MM-DD KST)
reference 외부 리소스 포인터 reference_*.md URL, 용도

실행 순서

1. 분류

입력 내용을 읽고 위 4가지 중 어디에 해당하는지 판단합니다. 애매하면 사용자에게 확인하세요. 복수 타입이면 각각 별도 파일로 분리합니다.

2. 파일명 제안

  • 소문자 + 언더스코어
  • 핵심 키워드 2~4개
  • 예시: feedback_use_testcontainers.md, project_mobile_freeze_2026q1.md

3. frontmatter + 본문 작성

기본 템플릿:

---
name: (한 줄 제목)
type: (분류한 타입)
description: (한 줄 요약)
---

(본문)

타입별 추가 요구사항:

  • feedback → **Why**: 와 **How to apply**: 두 줄 반드시 포함
  • project → **사실**: 에 YYYY-MM-DD KST 형식 절대 날짜 포함
  • user → 불릿 리스트로 간결하게
  • reference → URL과 "언제 참조하는가" 명시

4. 인덱스 업데이트

memory/MEMORY.md의 해당 섹션에 링크를 추가합니다:

  • user → ## 사용자 섹션
  • feedback → ## 기술 메모 섹션
  • project → ## 기술 메모 섹션 (진행 중인 것만)
  • reference → ## 외부 리소스 섹션

링크 형식: - [파일명.md](파일명.md) 한 줄 요약

5. 저장 전 확인

저장하기 전에 다음을 사용자에게 보여주고 승인을 받으세요:

  1. 제안된 파일 경로
  2. 작성된 frontmatter + 본문 전체
  3. MEMORY.md에 추가될 한 줄

사용자가 "OK" 또는 "저장해줘" 등으로 승인하면 실제로 Write합니다.

저장 금지 규칙

다음은 저장하지 마세요 (사용자에게 안내 후 스킵):

  • 코드에서 derivable 한 사실 (예: "함수 이름은 xxx")
  • git log/PR로 추적 가능한 변경사항
  • 일회성 정보 (이미 해결된 버그 등)
  • 민감 정보 (API 키, PII, 내부 URL 등)

민감 정보가 의심되면 경고하고 저장을 중단하세요.

특수 케이스

  • memory/ 디렉터리가 없으면 → /memory-init 실행을 권유
  • 동일 파일명이 이미 있으면 → 덮어쓸지, append할지, 다른 이름으로 할지 사용자에게 선택받기
  • 내용이 너무 짧거나 모호하면 → "더 구체적으로 적어달라"고 되묻기
Changes

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.

  1. 2mo ago First seen · 85 lines · 15 tokens per session scan A 463aa8ca7960

Subscribe to this mod's changes

memory-save is a command published in the GitHub repository gaebalai/claude-memory-plugin (2 stars, last pushed 6mo ago), licensed MIT. It adds 15 tokens to every session and 837 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Memory and context
Related

memory-save compared with commands from other repositories

memory-save next to 6 commands from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
memory-savegaebalai/claude-memory-pluginCommand26mo agoMITA15Claude Code
minutes-ideassilverstein/minutesCommand1.6kyesterdayMITA48Claude Code, OpenCode
learnReflexioAI/claude-smartCommand792todayApache-2.0A14Claude Code
memory-storeverygoodplugins/mcp-automemCommand66yesterdayMITA10Claude Code
cc-memoryTheLobbi/claudeCommand218d agoMITA0Claude Code
analyze-contextRohiRIK/OpenLtmCommand299d agoMITA39Claude Code
lians-recallLians-ai/LiansCommand175d agoApache-2.0A19Claude Code