# Lore

Lore gives AI agents shared memory, semantic recall, and linked facts across sessions. It fits projects where context must persist between agents.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 10
- Author: agentkitai
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/agentkitai/lore
- Updated: 2026-10-18
- HTML: https://affpapa.org/en/nevibecode/mcp/agentkitai-lore

## What it does

Lore stores knowledge through MCP, a REST API, or Python and TypeScript SDKs. It can recall semantically related memories, apply time-based decay, and redact PII. Its knowledge graph connects entities and facts, while bi-temporal queries let an agent inspect what was valid at a given point in time.

Hooks capture session context and inject relevant history into prompts without changes to the agent. The server also includes JSON and Markdown export, snapshots, a browser UI at `/ui/`, workspace isolation with RBAC, and Python entry-point plugins.

The recommended setup uses Docker Compose to start Postgres with pgvector and Lore at `http://localhost:8765`. This is a database-backed setup rather than a lightweight local-only MCP. The `lore bootstrap` command checks Python, Postgres, pgvector, Docker, migrations, and server health.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add lore -- uvx lore-sdk
```
uvx runs the command named after the package; if the README names another one, use uvx --from lore-sdk <command>

### Codex

**Platform CLI**
```bash
codex mcp add agentkitai-lore -- uvx lore-sdk
```
uvx runs the command named after the package; if the README names another one, use uvx --from lore-sdk <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "agentkitai-lore": {
            "command": "uvx",
            "args": [
                "lore-sdk"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from lore-sdk <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "agentkitai-lore": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "lore-sdk"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from lore-sdk <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add agentkitai-lore uvx lore-sdk
```
uvx runs the command named after the package; if the README names another one, use uvx --from lore-sdk <command>

## When to use it
- carry project context between agent sessions
- retrieve facts connected through a knowledge graph
- inspect which facts were valid on a past date
- export stored knowledge as JSON or Markdown
- review graph connections by risk or confidence

## Example prompts
- Save the project’s API rate-limit decision and link it to related knowledge.
- Find memories about the Postgres setup and show connected entities.
- Tell me which project facts were valid on the date I provide.

## Similar
- [pindoc](https://affpapa.org/en/nevibecode/mcp/pindoc.md) — MCP server · ★ 12
- [Memory Vault](https://affpapa.org/en/nevibecode/mcp/memory-vault.md) — MCP server · ★ 65
- [Open Brain](https://affpapa.org/en/nevibecode/mcp/open-brain.md) — MCP server · ★ 8
- [postgres](https://affpapa.org/en/nevibecode/skills/postgres.md) — skill · 1400 installs
- [pgvector-semantic-search](https://affpapa.org/en/nevibecode/skills/pgvector-semantic-search.md) — skill · 938 installs
- [postgres-skills](https://affpapa.org/en/nevibecode/plugins/postgres-skills.md) — plugin · ★ 42

## Other versions with the same name
- [lore · jordanhindo](https://affpapa.org/en/nevibecode/mcp/lore.md) — ★ 16

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md