Skip to main content
Glama
Unified Memory MCP

Unified Memory MCP

by yksanjo

Unified Memory MCP

Local, explicit memory vault exposed as an MCP server. It is meant to give Codex, ChatGPT Apps, and your own agent clients a shared memory layer without relying on private product memory crossing surfaces automatically.

What It Does

  • Stores memories in ~/.unified-memory-mcp/memories.json.

  • Writes a minimal audit log to ~/.unified-memory-mcp/events.jsonl.

  • Exposes MCP tools over stdio.

  • Hides sensitive: true memories from default read/search/list calls.

  • Keeps the storage format human-readable and easy to back up or inspect.

Related MCP server: Codex Memory

Tools

  • remember: create a memory.

  • search_memory: search active memories by text and filters.

  • list_memories: list recent memories.

  • get_memory: fetch a memory by id.

  • update_memory: patch an existing memory.

  • forget_memory: hard-delete a memory.

  • memory_stats: show vault paths and aggregate counts.

Memory Shape

{
  "id": "mem_...",
  "type": "preference",
  "scope": "personal",
  "namespace": "default",
  "content": "Prefer pnpm over npm for JavaScript projects.",
  "tags": ["javascript", "package-manager"],
  "source": "manual",
  "confidence": 0.8,
  "sensitive": false,
  "created_at": "2026-07-04T00:00:00.000Z",
  "updated_at": "2026-07-04T00:00:00.000Z",
  "expires_at": null,
  "metadata": {}
}

Allowed type values: preference, project, workflow, fact, decision, context, rule.

Allowed scope values: personal, company, repo, project, global.

Local Usage

cd /Users/yoshikondo/unified-memory-mcp
npm test
npm run smoke
npm run smoke:http

Run the MCP server directly:

MEMORY_MCP_HOME=/Users/yoshikondo/.unified-memory-mcp node /Users/yoshikondo/unified-memory-mcp/src/server.js

Run the optional Streamable HTTP MCP server:

MEMORY_MCP_HOME=/Users/yoshikondo/.unified-memory-mcp \
MEMORY_MCP_HOST=127.0.0.1 \
MEMORY_MCP_PORT=3199 \
npm run http

The HTTP endpoint is http://127.0.0.1:3199/mcp; health is available at http://127.0.0.1:3199/health.

If you bind to anything other than localhost, set MEMORY_MCP_TOKEN and put the service behind HTTPS before connecting remote clients.

Codex Setup

Add this to ~/.codex/config.toml:

[mcp_servers.unified_memory]
command = "node"
args = ["/Users/yoshikondo/unified-memory-mcp/src/server.js"]
startup_timeout_sec = 10
tool_timeout_sec = 30

[mcp_servers.unified_memory.env]
MEMORY_MCP_HOME = "/Users/yoshikondo/.unified-memory-mcp"

Restart Codex after changing MCP config.

Privacy Notes

Do not store secrets, credentials, private keys, API tokens, or raw sensitive personal data. Mark entries sensitive: true only as an extra visibility guard; it is not encryption.

This is separate from built-in ChatGPT memory and Codex memory. It is an explicit source of truth that clients can connect to through MCP.

For ChatGPT, this is the backend MCP surface. You still need a ChatGPT App or remote MCP deployment with HTTPS and appropriate auth before ChatGPT web can use it.

Available Tools

7 tools
forget_memoryForget MemoryB
Destructive

Hard-delete a memory entry from the local vault.

Inputs: id (required), reason.

get_memoryGet MemoryA
Read-onlyIdempotent

Fetch one active memory by id. Sensitive memories are hidden unless include_sensitive is true.

Inputs: id (required), include_sensitive.

list_memoriesList MemoriesB
Read-onlyIdempotent

List recent active memories with optional filters. Sensitive memories are excluded by default.

Inputs: tags, type, limit, scope, +3 more.

memory_statsMemory StatsB
Read-onlyIdempotent

Return vault paths and aggregate counts.

Inputs: no named parameters.

rememberRememberB

Create a durable memory entry in the local vault.

Inputs: content (required), tags, type, scope, +6 more.

search_memorySearch MemoryA
Read-onlyIdempotent

Search active memory entries by query and filters. Sensitive memories are excluded by default.

Inputs: tags, type, limit, query, +4 more.

update_memoryUpdate MemoryB

Patch an existing memory entry by id.

Inputs: id (required), tags, type, scope, +7 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 7 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: stats, create, search, list, get, update, delete. No overlaps, and sensitive memory handling is consistently described.

    Naming Consistency3/5

    Naming is inconsistent: most tools follow verb_noun pattern (search_memory, list_memories, etc.), but 'memory_stats' is noun_verb and 'remember' is just a verb, breaking the pattern.

    Tool Count5/5

    7 tools cover essential memory operations (CRUD, search, list, stats) without being excessive or insufficient.

    Completeness5/5

    Full CRUD lifecycle plus search and statistics. Sensitive memory handling is integrated. No obvious gaps for a memory vault.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      Provides a local, persistent long-term memory service for MCP-compatible AI agents, enabling them to store, search, and recall information across sessions.
      1
      GPL 3.0
    • A
      license
      B
      quality
      D
      maintenance
      Local Markdown-backed memory tools for Codex and other MCP-capable agents. Exposes durable agent knowledge via CLI and MCP server.
      5
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Offers MCP tools for managing a local AI memory vault, including remembering, searching, context packing, deletion, and verification of receipts.
      87 npm
      1
      Apache 2.0
    • A
      license
      Not graded
      quality
      C
      maintenance
      MCP server that provides agentic memory management for markdown vaults, enabling hybrid search, governed writing, and maintenance of episodic, semantic, procedural, and working memories for LLM agents.
      MIT