nmg-memory

nmg-memory is a skill for Claude Code, Codex from wefio/NodeMemoryGraph. It costs 73 tokens per session (2,315 once invoked), scanned A, original, MIT.

A skill for using NMG as durable memory or as a temporary shared task board between agents. It covers remembering user facts, decisions, project state and reusable experience.

In plain words
What is it for?
Use it to recall or save attributable information, manage temporary multi-agent coordination and preserve useful project knowledge.
Why use it?
It helps an agent find relevant prior context and coordinate work without treating memories or temporary notes as automatically reliable.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/wefio/nodememorygraph/nmg-memory
Any agent
npx skills add wefio/NodeMemoryGraph --skill nmg-memory
Clone the repo
git clone --depth 1 https://github.com/wefio/NodeMemoryGraph

Made for: Claude Code, Codex.

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 nmg-memory

README.md
[![agentmods](https://agentmods.dev/badge/skills/wefio/nodememorygraph/nmg-memory.svg)](https://agentmods.dev/skills/wefio/nodememorygraph/nmg-memory)
Your own site
<a href="https://agentmods.dev/skills/wefio/nodememorygraph/nmg-memory"><img src="https://agentmods.dev/badge/skills/wefio/nodememorygraph/nmg-memory.svg" alt="Measured on agentmods" height="20"></a>
Per session 73 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,315 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00073 $0.02315
Opus 5 $0.00036 $0.01157
Sonnet 5 $0.00015 $0.00463
Haiku 4.5 $0.00007 $0.00231

Measured 4d ago against content hash 8b8306e0db6a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

nmg-memory 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 4d 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.

skills/nmg-memory/SKILL.md · 206 lines

How it starts

The opening of the file, as written. The whole thing — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.

NMG Memory

Treat this file as the quick-start card, not a document to reread every turn. Once the workflow is known, use it directly. Read a reference only after forgetting an operation or encountering its named special case.

Responsibilities and boundaries

  • Use NMG to recall and save durable information; do not treat it as the final judge of truth, currency, relevance, or evidence completeness.
  • Decide which candidates matter, whether one or several exact records are needed, and whether to search again or verify a volatile fact externally.
  • Accept no useful memory as a valid outcome.
  • Save only attributable durable information. Keep secrets, transient content, unconfirmed Assistant proposals, and unsupported guesses out of memory.
  • When temporary cross-Agent coordination is enabled, use the task board rather than LTG or Markdown files. Blackboard entries expire and never become durable memory unless an Agent separately calls nmg remember with attributable evidence. Ordinary single-Agent memory use does not require the board.
  • Let ordinary work produce natural improvement evidence when controller shadow collection is enabled. Record only outcomes that are directly observable; an uncorrected answer, silence, retrieval, or answer reuse remains unknown.

Normal workflow

  1. Check nmg daemon status --json and inspect both running and compatible.

  2. If it is not running, run nmg daemon start --json and remember that this Agent invocation owns the daemon. If it is running but compatible=false, do not reuse or automatically replace it: report that a coordinated nmg daemon restart is required. A shared daemon may still serve another active Agent, so only its owner or the user may choose the safe restart point.

  3. Before answering a history-dependent question, run:

    nmg search "<focused recall query>" --project-dir . --limit 8 --max-tier 1 --compact-json
    

    Narrow the scope when the store is large or the topic is specific: --node "<name>", --scope project=NAME, --source-actor user, --include-historical, --max-tier 2, --graph-hops 2.

Read the full file on GitHub · 206 lines

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. 4d ago First seen · 206 lines · 73 tokens per session scan A 8b8306e0db6a

Subscribe to this mod's changes

nmg-memory is a skill published in the GitHub repository wefio/NodeMemoryGraph (0 stars, last pushed yesterday), licensed MIT. It adds 73 tokens to every session and 2,315 once invoked, about $0.0004 per session on Opus 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.

Related

Other skills, from other repositories

skills-constitution

当 Agent 接到专业任务(编码/爬虫/文件操作/API调用/数据分析/文档/部署/推送等)时,强制先查记忆层和技能索引,有匹配必用、无匹配必搜、答复时自动推荐(排除已装)。用于防止 Agent 跳过技能直接硬扛通用能力。跨平台通用(WorkBuddy/Claude/ChatGPT/Cursor/Gemini 等 20+ 框架)。完整版本史见 CHANGELOG.md。.

jiabaobei/skills-constitution · 119 tokens

add-memory-system

Research and integrate a new agent-memory repository into Agent Memory Atlas. Use when asked to add, analyze, catalog, or compare a memory system in this repository, including its commit-pinned system report, comparative overview coverage, homepage card, generated site, and validation.

neoneye/agent-memory-atlas · 57 tokens

reanalyze-memory-system

Re-read a repository Agent Memory Atlas already has a report for, at a newer commit, and fold what changed into the existing report. Use when asked to re-analyze, re-read, re-pin, refresh, or check for updates on a system already in the atlas, when an upstream project reports a fix, or when a freshness check flags a…

neoneye/agent-memory-atlas · 78 tokens

use-the-atlas

Design, review, or build a memory system for some other product, using the Agent Memory Atlas as the reference. Use when asked to add memory, persistence, or recall to a repository, to review an existing memory implementation against the atlas, or to decide which memory patterns a product actually needs. Runs in one…

neoneye/agent-memory-atlas · 92 tokens

engrama-onboard

Build a personalized Engrama memory graph for any user through a conversational interview. Use this skill whenever someone says "set up Engrama", "create my graph", "onboard", "build my profile", "configure Engrama for me", or asks how to start using Engrama. Also use when the user mentions they have a different role…

scops/engrama · 104 tokens

growmos

Use the repository's living knowledge graph (growmos, in .growmos/) as shared memory. Trigger when the user asks about how parts of the codebase relate, why a decision was made, who owns what, what depends on what; when you finish a meaningful piece of development and should record it; when growmos next reports…

codician-team/growmos · 97 tokens