# Total Agent Memory

Stores decisions, bugs, and project facts in a local SQLite database across agent sessions. Keeps context intact so you do not repeat project architecture details.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 73
- Author: vbcherepanov
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/vbcherepanov/total-agent-memory
- Updated: 2026-10-14
- HTML: https://affpapa.org/en/nevibecode/mcp/total-agent-memory

## What it does

The server saves architectural decisions, session summaries, and bug records into a local SQLite store in `~/.tam/`. Retrieval fuses BM25 indexing, dense vector representations, and a knowledge graph. Default searches do not call external models. Search results stay deterministic.

Running this tool requires Python 3.11 or newer. Without Python present, running through `uvx` fails immediately. An optional cross-encoder reranker requires extra PyTorch packages via `total-agent-memory[rerank]`. In addition, stored facts track valid timeframes, so newer records cleanly supersede older values.

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

## Install

### Claude Code (native platform)

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

### Codex

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

### Cursor

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

### GitHub Copilot

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

### Gemini CLI

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

## When to use it
- preserve architectural decisions across separate coding sessions
- stop repeating tech stack conventions to the assistant
- trace fact validity over time using temporal knowledge graphs
- recall past bug solutions and debugging summaries locally

## Example prompts
- Remember that we picked PostgreSQL for billing due to row-level security
- Which database did we select for the billing service and why?
- Check memory for the solution we used on the previous auth bug

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 386989
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33918
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81115
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47392
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43872

---
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