# Lithtrix — Identity, Memory & Trust for AI Agents

Gives an agent self-registration, an Ed25519 passport, shared memory and web search behind one ltx_ key.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 1
- Author: lithtrix
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/lithtrix/lithtrix-mcp
- Updated: 2026-06-20
- HTML: https://affpapa.org/en/nevibecode/mcp/lithtrix-identity-memory-trust-for-ai-agents

## What it does

The server wraps four layers around an agent: memory, identity, reputation and the shared Commons. An agent registers itself with `lithtrix_register`, gets an `ltx_` key and an Ed25519 passport, then works with memory, blobs, search and feedback without a human setting anything up. MIRC stays free; tools like `lithtrix_browse` and `lithtrix_search` are metered per call.

What you get is a set of MCP tools. `lithtrix_memory_set` and `lithtrix_memory_get` handle keys, `lithtrix_memory_search` does semantic lookup, blob tools upload, download and parse files, and `lithtrix_feedback` scores earlier results. Version `0.20.1` brought `lithtrix_spawn`, `lithtrix_delegate`, `lithtrix_trace_append` and `lithtrix_trace_get` — swarm primitives where the delegation contract is signed on the client. Signing needs `LITHTRIX_PASSPORT_MASTER_SEED` or `LITHTRIX_PASSPORT_PRIVATE_KEY`. Without either, delegation will not run.

The key is read from `process.env.LITHTRIX_API_KEY` only, so keep it out of the tool definition and pass it through your host's credential store or a secrets manager. Registration fails unless `agree_to_terms` is `true`. New agents start with $5 in trial credits and no card.

## Requirements
- Needs: Node.js (commands run via npx)
- Environment variables: `LITHTRIX_API_KEY` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add lithtrix-mcp -e LITHTRIX_API_KEY=<LITHTRIX_API_KEY> -- npx -y lithtrix-mcp
```
Replace <…> with your own values

### Codex

**Platform CLI**
```bash
codex mcp add lithtrix-identity-memory-trust-for-ai-agents --env LITHTRIX_API_KEY=<LITHTRIX_API_KEY> -- npx -y lithtrix-mcp
```
Replace <…> with your own values

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "lithtrix-identity-memory-trust-for-ai-agents": {
            "command": "npx",
            "args": [
                "-y",
                "lithtrix-mcp"
            ],
            "env": {
                "LITHTRIX_API_KEY": "<LITHTRIX_API_KEY>"
            }
        }
    }
}
```
Replace <…> with your own values

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "lithtrix-identity-memory-trust-for-ai-agents": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "lithtrix-mcp"
            ],
            "env": {
                "LITHTRIX_API_KEY": "<LITHTRIX_API_KEY>"
            }
        }
    }
}
```
Replace <…> with your own values

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add -e LITHTRIX_API_KEY=<LITHTRIX_API_KEY> lithtrix-identity-memory-trust-for-ai-agents npx -y lithtrix-mcp
```
Replace <…> with your own values

## When to use it
- give an agent an identity and passport without manual signup
- store and fetch agent memory by key
- run semantic search over stored memory
- upload and parse files as blobs
- rate earlier search results and flag wrong ones

## Example prompts
- Register a new agent with `lithtrix_register` and save the key
- Store a short task description in memory under the key `project-notes`
- Search memory for anything tied to the last release

## Similar
- [atxp-memory](https://affpapa.org/en/nevibecode/skills/atxp-memory.md) — skill · 735 installs
- [use-link-identity](https://affpapa.org/en/nevibecode/skills/use-link-identity.md) — skill · 9500 installs
- [verified-agent-identity](https://affpapa.org/en/nevibecode/skills/verified-agent-identity.md) — skill · 8200 installs
- [Synapse Layer — Trust Infrastructure for AI Agents](https://affpapa.org/en/nevibecode/mcp/synapse-layer-trust-infrastructure-for-ai-agents.md) — MCP server · ★ 13
- [azure-identity-ts](https://affpapa.org/en/nevibecode/skills/azure-identity-ts.md) — skill · 1 install
- [deep-agents-memory](https://affpapa.org/en/nevibecode/skills/deep-agents-memory.md) — skill · 19000 installs

---
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