# KuzuMem

> Use this tool when you need to manage complex software systems with multiple AI agents, requiring structured development phases and unified tools for architectural decisions, component dependencies, and context logging. KuzuMem provides distributed graph memory bank storage, enforcing workflow states and offering features like graph algorithms, cycle detection, and full-text search. It is ideal for teams needing to maintain architectural context and decision history across sessions, with specialized agent modes and AI-powered memory optimization.

Canonical page: https://skillsregistry.net/skills/jakedismo-kuzumem  
JSON: https://api.skillsregistry.net/v1/skills/jakedismo-kuzumem

## Description

KuzuMem-MCP provides distributed graph memory bank storage for coding agents using KuzuDB, built with TypeScript and featuring a comprehensive workflow state machine that enforces structured development phases (ANALYZE, BLUEPRINT, CONSTRUCT, VALIDATE, ROLLBACK). The implementation offers 12 unified tools for managing architectural decisions, component dependencies, file tracking, and context logging, with specialized agent modes for different development roles (architect, code, debug) and AI-powered memory optimization using OpenAI and Anthropic models. Features include graph algorithms for dependency analysis, cycle detection, full-text search across entities, and persistent memory across sessions, making it valuable for teams building complex software systems where maintaining architectural context and decision history across multiple AI agents is critical for long-term project success.

## Trust

- **Trust score (0–1):** 0.61
- **Verification tier:** scanned
- **Last scanned:** 2026-09-19

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** ai-ml
- **Updated:** 2026-09-19

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/jakedismo-kuzumem)
- **Repository:** <https://github.com/jakedismo/kuzumem-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "jakedismo-kuzumem"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/jakedismo-kuzumem` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/jakedismo-kuzumem/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt