# Synapse Layer — Trust Infrastructure for AI Agents

Persistent memory for AI agents: entries are AES-256-GCM encrypted, recall is semantic, access goes over MCP. Useful when your agent drops context between sessions.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 13
- Author: SynapseLayer
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/SynapseLayer/synapse-layer
- Updated: 2026-10-22
- HTML: https://affpapa.org/en/nevibecode/mcp/synapse-layer-trust-infrastructure-for-ai-agents

## What it does

Synapse Layer keeps facts, preferences and past decisions alive across agent runs. Entries sit encrypted at rest with AES-256-GCM, a fresh IV per operation and HMAC-SHA-256 integrity checks, while retrieval runs on pgvector HNSW — by meaning rather than exact wording. Every entry carries a Trust Quotient, so results come back ranked by confidence instead of recency alone.

The server exposes 13 MCP tools: `recall`, `save_to_synapse`, `process_text`, `search`, `health_check`, `initialize_context`, `save_memory`, `store_memory`, `recall_memory`, `list_memories`, `memory_feedback`, `neural_handover`, `slo_report`. Through them an agent writes context, reads it back, leaves feedback and hands work to another agent. Memory is shared, so what one client stores another can pull.

You need a Connect Token, issued at forge.synapselayer.org under Connect. It travels in the `x-connect-token` header, never in a URL or a log line. Two routes exist: the hosted Forge API at `https://forge.synapselayer.org/api/mcp`, or the `synapse-layer` Python SDK with the token passed in code. Without a token the server answers nothing.

## Requirements
- Remote server: `https://forge.synapselayer.org/api/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http synapse-layer https://forge.synapselayer.org/api/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add synapse-layer-trust-infrastructure-for-ai-agents --url https://forge.synapselayer.org/api/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "synapse-layer-trust-infrastructure-for-ai-agents": {
            "url": "https://forge.synapselayer.org/api/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "synapse-layer-trust-infrastructure-for-ai-agents": {
            "type": "http",
            "url": "https://forge.synapselayer.org/api/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http synapse-layer-trust-infrastructure-for-ai-agents https://forge.synapselayer.org/api/mcp
```

## When to use it
- keep user preferences across agent sessions
- find a stored fact by meaning, not by exact word
- move context from one agent to another
- keep memory encrypted on disk
- hand unfinished work to a different agent

## Example prompts
- Save to memory that I prefer dark mode and short answers
- Search memory for everything we decided about auth
- Hand the current context to the next agent with `neural_handover`

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 392262
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33989
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81163
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119231
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47577
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43902

---
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