# AI Memory

> Use this tool when you need to manage contextual knowledge across sessions for AI agents, solving problems of persistent memory and knowledge management in conversational systems. It takes in various data types and outputs stored, retrieved, and managed knowledge, utilizing interfaces like PostgreSQL and Redis for efficient storage and caching. Ideal for use cases like multi-session conversational agents and knowledge management systems, AI Memory enables intelligent caching, clustering, and compression for effective semantic memory management.

Canonical page: https://skillsregistry.net/skills/ai-memory  
JSON: https://api.skillsregistry.net/v1/skills/ai-memory

## Description

Production-ready MCP server for semantic memory management that enables AI agents to store, retrieve, and manage contextual knowledge across sessions using PostgreSQL with pgvector for vector similarity search and local Transformers.js embeddings. Features intelligent caching with Redis fallback, multi-agent support through user context isolation, memory relationships for connected knowledge graphs, automatic clustering with DBSCAN algorithm, smart compression for large content, and background job processing with BullMQ for async embedding generation and batch operations. Built with TypeScript and Kysely ORM for type safety, it provides soft deletes, input sanitization, token-efficient responses, and flexible embedding dimensions, making it ideal for AI applications requiring persistent memory, knowledge management systems, and multi-session conversational agents.

## Trust

- **Trust score (0–1):** 0.57
- **Verification tier:** scanned
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** database
- **Updated:** 2026-09-28

## Source

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

## 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": "ai-memory"
    }
  }
}
```

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