Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
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.
[](https://agentmods.dev/instructions/adamkwhite/universal-memory-mcp/claude-md)<a href="https://agentmods.dev/instructions/adamkwhite/universal-memory-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/adamkwhite/universal-memory-mcp/claude-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 5.1 | $0.13068 | $0.13068 |
| Opus 5 | $0.06534 | $0.06534 |
| Sonnet 5 | $0.02614 | $0.02614 |
| Haiku 4.5 | $0.01307 | $0.01307 |
Grade D, and why
universal-memory-mcp CLAUDE.md scanned grade D with 3 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 6d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
`curl -s "https://sonarcloud.io/api/issues/search?componentKeys=adamkwhite_claude-memory-mcp&resolved=false" | python3 -c "import json,sys; print(json.load(sys.stdin)['total'])"`. Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
and Codex's `~/.codex/config.toml` — dry-run by default, `--apply` writes after a timestamped Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`curl -s "https://sonarcloud.io/api/issues/search?componentKeys=adamkwhite_claude-memory-mcp&resolved=false" | python3 -c "import json,sys; print(json.load(sys.stdin)['total'])"`. How it starts
The opening of the file, as written. The whole thing — 906 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Universal Memory MCP - Universal AI Memory System
Rule for editing this file: never restate an enumeration. MCP tool inventories, test counts,
required-check lists, file:line citations, table/column rosters — name the code or command that is
true by construction (grep '@mcp.tool()' -A2 src/universal_memory_mcp/server_fastmcp.py, pytest -q | tail -1,
gh api repos/adamkwhite/universal-memory-mcp/rulesets/5957219) and let the reader run it. A prose
copy of an enumerable fact rots while the thing it copies moves, and it rots while still reading as
authoritative — which is worse than being absent, because nobody re-checks a confident sentence.
Adopted from job-agent's CLAUDE.md:7, where it is measurably working: a drift sweep of that file
on 2026-08-16 found four stale claims and zero enumeration drift.
It cannot cover everything, and knowing where it stops matters. A behavioural claim — "this script
aborts on failure", "that context manager doesn't close the connection" — has no pointer-to-code
form. You either assert it or lose the knowledge. Those are what the /EndSession CLAUDE.md drift
sweep exists for. Prevention for enumerable facts, sweep for behavioural ones.
Project Overview
Claude Memory MCP is a universal conversation memory system that provides persistent storage and intelligent search across multiple AI platforms. Originally designed for Claude, it now supports ChatGPT, Cursor AI, and custom formats through an extensible architecture.
Current Status (August 20, 2026)
Branch: main
Recent Work: SonarCloud debt cleared to zero (#232, via #238–#240/#244/#247) and
scripts/switch_mcp_config.py added to repoint MCP configs at the published console script
(#245/#246) — the #225 package move broke every config that launched the server as a loose file.
Before that: published to PyPI as universal-memory-mcp (#230, tag v0.1.0; now 0.1.2),
which required #225 — the wheel had been shipping without the application. See todos.md.
Test Coverage: run pytest -q | tail -1 for the local count and read the Tests (Windows) job for the CI one — both were restated here and both had already drifted. Windows carries a small number of deliberate POSIX-only skips (see the quality-gate section). Coverage: SonarCloud is authoritative, not a local .coverage file; ≥80% required on new code.
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.
- 6d ago First seen · 906 lines · 13,068 tokens per session scan D 32b859be8933
universal-memory-mcp CLAUDE.md is an instructions file published in the GitHub repository adamkwhite/universal-memory-mcp (3 stars, last pushed 5d ago), licensed MIT. It adds 13,068 tokens to every session, about $0.0653 per session on Opus 5. A static security scan graded it D with 3 findings (downloads and executes remote code, reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
aeon CLAUDE.md
Claude Code instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
cognirepo CLAUDE.md
Claude Code instructions for ashlesh-t/cognirepo, covering claude.md, key rules, session start sequence (run in this order), behavioral confirmation rule and personas (cognirepo-402, cognirepo-403).
wayland-core copilot-instructions.md
Copilot instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
mcp-structured-memory CLAUDE.md
Claude Code instructions for nmeierpolys/mcp-structured-memory, a project described as: Structured Memory MCP Server.
inkwell-memory CLAUDE.md
Instructions for veronchenko/inkwell-memory, covering claude.md — inkwellmemory, layout, multi-tenant mode (inkwellmultitenant=1), conventions and testing.
RNR-Enhanced-Cognee AGENTS.md
AGENTS.md instructions for vincentspereira/RNR-Enhanced-Cognee, covering rnr enhanced cognee implementation for codex, critical requirements, 1. ascii-only output (no unicode encoding), 2. dynamic categories (no hardcoded categories) and 3. standard memory mcp interface.