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.
npx agentmods add instructions/verygoodplugins/mcp-automem/agents-mdgit clone --depth 1 https://github.com/verygoodplugins/mcp-automemWrote 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/instructions/verygoodplugins/mcp-automem/agents-md)<a href="https://agentmods.dev/instructions/verygoodplugins/mcp-automem/agents-md"><img src="https://agentmods.dev/badge/instructions/verygoodplugins/mcp-automem/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.08172 | $0.08172 |
| Opus 5 | $0.04086 | $0.04086 |
| Sonnet 5 | $0.01634 | $0.01634 |
| Haiku 4.5 | $0.00817 | $0.00817 |
Grade B, and why
mcp-automem AGENTS.md scanned grade B with 1 finding 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **CLI installer (settings-level alternative + migration path):** the `claude-code` command writes hook scripts into `~/.claude/hooks/` and merges hook registrations plus exactly the six `mcp__memory__*` permissions int How it starts
The opening of the file, as written. The whole thing — 536 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This file provides guidance to coding agents (Claude Code, Cursor, Codex, etc.) when working with code in this repository. It is the single source of truth; CLAUDE.md imports it via @AGENTS.md.
Migration note: if you previously had a local, gitignored AGENTS.md, delete it before pulling (or run git clean -f AGENTS.md) to avoid checkout/merge errors.
Project Overview
MCP AutoMem is an MCP (Model Context Protocol) server that bridges AI assistants like Claude with the AutoMem memory service. It enables AI to store, recall, and associate memories using FalkorDB (graph database) and Qdrant (vector search).
Core Purpose:
- Translate MCP tool calls into AutoMem API requests
- Provide memory management for AI assistants (storage, hybrid search, relationships)
- Support Claude Code integration with session-recall and opt-in storage-nudge hooks
Build & Development
# Build TypeScript to dist/
npm run build
# Development with hot-reload
npm run dev
# Test server help output
npm test
# Publish preparation
npm run prepublishOnly # Runs build automatically
# Formatting (CI runs format:check before lint)
npm run format
npm run format:check
Prettier owns src/, tests/, scripts/, and repo config files. .prettierignore excludes
templates/, plugins/, and skills/ because those are regenerated by the memory-policy
sync on every build — formatting them would fight the generator. Markdown is excluded too:
tests/docs/integration-coverage.test.ts asserts against README/INSTALLATION prose and table
sections, so reflowing them risks silent coverage breakage.
Commit Standards (Required)
This repo uses Conventional Commits so Release Please can generate releases reliably.
Accepted examples:
fix: prevent stdout corruption in stdio mode
feat: add cursor setup command
chore: update dependencies
docs: clarify Claude Desktop config
Notes:
- PR titles must be Conventional Commit format because we squash-merge and the PR title becomes the merge commit message.
- A git
commit-msghook (Husky + Commitlint) is included to catch mistakes locally, and CI enforces PR titles.
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.
- 4d ago First seen · 536 lines · 8,172 tokens per session scan B b739fff23c9f
mcp-automem AGENTS.md is an instructions file published in the GitHub repository verygoodplugins/mcp-automem (64 stars, last pushed today), licensed MIT. It adds 8,172 tokens to every session, about $0.0409 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
automem CLAUDE.md
Instructions for verygoodplugins/automem, covering claude.md, project overview, development commands, setup environment and development.
faulkner-db AGENTS.md
Instructions for Platano78/faulkner-db, covering faulkner db — agents.md (router / the map), what this is, areas (route by task — load inputs, skip the rest), verbs and naming conventions (locate files, don't grep blindly).
tinkerpop AGENTS.md
AGENTS.md instructions for apache/tinkerpop, covering agents.md, primary guidance: agent skills, planning and memory: beads, canonical documentation and licensing and provenance.
automem AGENTS.md
Instructions for verygoodplugins/automem, covering repository guidelines, project structure & modules, build, test, and development, coding style & naming and testing guidelines.
codex-multi-auth AGENTS.md
AGENTS.md instructions for ndycode/codex-multi-auth, covering project knowledge base, structure, where to look, conventions and anti-patterns.
kglite CLAUDE.md
Claude Code instructions for kkollsga/kglite, covering kglite — claude code conventions, build & test, architecture, the boundary principle (wrappers vs core) — summary and in-memory is the core product.